Directories
-----------
- fltk - FLTK GUI programs(s)
- octave - Octave scripts used for visualising internal signals
- during development
- script - shell scripts for playing and converting raw files
- src - C source code
- raw - speech files in raw format (16 bits signed linear 8 kHz)
- unittest - unit test source code
- voicing - hand-estimated voicing files, used for development
- wav - speech files in wave file format
- win32 - Support for building Windows DLL version of Codec 2 and FDMDV libraries
+ fltk - FLTK GUI programs(s)
+ octave - Octave scripts used for visualising internal signals
+ during development
+ portaudio - Portaudio test programs
+ script - shell scripts for playing and converting raw files
+ src - C source code
+ raw - speech files in raw format (16 bits signed linear 8 kHz)
+ unittest - unit test source code
+ voicing - hand-estimated voicing files, used for development
+ wav - speech files in wave file format
+ win32 - Support for building Windows DLL version of Codec 2 and FDMDV libraries
TODO
----
-[ ] Get win32/Makefile integarted into Automake system, such that if
+[ ] Get win32/Makefile integrated into Automake system, such that if
i586-mingw32msvc exists the Win32 code gets automatically built.
+[ ] Same for fltk & portaudio, build these conditionally if libs exist
\ No newline at end of file
The ouput of the demod is codec voice data so it's OK if we miss or
repeat a frame every now and again.
+ TODOs:
+
+ + this might work with arbitrary input and output buffer lengths,
+ 0,1, or 2 only apply if we are inputting the nominal number of
+ samples on every call.
+
+ + so the I/O buffer sizes might not matter, as long as they of
+ reasonable size (say twice the nominal size).
+
\*------------------------------------------------------------------*/
void per_frame_rx_processing(short output_buf[], /* output buf of decoded speech samples */
--- /dev/null
+% lspwarp.m
+% David Rowe Sep 2012
+%
+% Experimenting with non-linear LSP frequency axis for LSP quantisation
+% Plots a scaled mel axis.
+
+1;
+
+function mel = freq2mel(f)
+ mel = 70*log10(1 + f/700);
+endfunction
+
+function freq = mel2freq(m)
+ freq = 700*(10 ^ (m/70) - 1);
+endfunction
+
+x = []; y = [];
+
+for freq = 100:25:4000
+ mel = freq2mel(freq);
+ x = [x freq];
+ y = [y mel];
+end
+
+plot(x,y)
+grid
+
+mel_start = floor(freq2mel(100));
+mel_end = floor(freq2mel(4000));
+
+x = []; y = [];
+for mel=mel_start:mel_end
+ freq = mel2freq(mel);
+ x = [x freq];
+ y = [y mel];
+end
+
+hold on;
+plot(x,y, '+')
+hold off;
% switch some stuff off to unclutter display
+ plot_Am = 0;
+ plot_Amq = 0;
+ plot_err = 0;
plot_lsp = 0;
plot_snr = 0;
plot_vsnr = 0;
- plot_sw = 1;
+ plot_sw = 0;
plot_pw = 1;
plot_pwb = 1;
plot_rw = 1;
axis([1 length(s) -20000 20000]);
figure(2);
+ clf;
Wo = model(f,1);
L = model(f,2);
Am = model(f,3:(L+2));
- plot((1:L)*Wo*4000/pi, 20*log10(Am),";Am;r");
+ if plot_Am
+ plot((1:L)*Wo*4000/pi, 20*log10(Am),";Am;r");
+ end
axis([1 4000 -10 80]);
hold on;
if plot_sw
if (file_in_path(".",modelq_name))
Amq = modelq(f,3:(L+2));
- plot((1:L)*Wo*4000/pi, 20*log10(Amq),";Amq;g" );
+ if plot_Amq
+ plot((1:L)*Wo*4000/pi, 20*log10(Amq),";Amq;g" );
+ end
if (file_in_path(".",pwb_name) && plot_pwb)
- plot((0:255)*4000/256, 10*log10(Pwb(f,:)),";Pwb;c");
+ plot((0:255)*4000/256, 10*log10(Pwb(f,:)),";Pwb;r");
endif
if (file_in_path(".",rw_name) && plot_rw)
- plot((0:255)*4000/256, 10*log10(Rw(f,:)),";Rw;c");
+ plot((0:255)*4000/256, 10*log10(Rw(f,:)),";Rw;b");
endif
if (file_in_path(".",pw_name) && plot_pw)
- plot((0:255)*4000/256, 10*log10(Pw(f,:)),";Pw;c.");
+ plot((0:255)*4000/256, 10*log10(Pw(f,:)),";Pw;g.");
endif
signal = Am * Am';
noise = (Am-Amq) * (Am-Amq)';
snr1 = 10*log10(signal/noise);
Am_err_label = sprintf(";Am error SNR %4.2f dB;m",snr1);
- plot((1:L)*Wo*4000/pi, 20*log10(Amq) - 20*log10(Am), Am_err_label);
+ if plot_err
+ plot((1:L)*Wo*4000/pi, 20*log10(Amq) - 20*log10(Am), Am_err_label);
+ end
endif
A2(i,:) = -20*log10(abs(fft(ak2(i,:),Ndft)));
sd(i) = sum((A1(i,:) - A2(i,:)).^2)/Ndft;
end
+ printf("sd av %3.2f dB*dB\n", sum(sd)/frames);
figure(1);
clf;
lsp1 = load(lsp1_filename);
lsp2 = load(lsp2_filename);
+ weights_filename = sprintf("%s_weights.txt", dump_file_prefix);
+ if file_in_path(".",weights_filename)
+ weights = load(weights_filename);
+ end
+
k = ' ';
do
figure(2);
axis([1 4000 -20 40]);
hold on;
plot((1:Ndft/2)*4000/(Ndft/2), A2(f,1:(Ndft/2)),";A2;");
-
+ if file_in_path(".",weights_filename)
+ plot(lsp1(f,:)*4000/pi, weights(f,:),";weights;g+");
+ end
+
for l=1:10
plot([lsp1(f,l)*4000/pi lsp1(f,l)*4000/pi], [0 -10], 'r');
plot([lsp2(f,l)*4000/pi lsp2(f,l)*4000/pi], [-10 -20], 'b');
LIBS = -lm -lportaudio -pthread
SRC = ../src/fdmdv.c ../src/kiss_fft.c
-all: pa_rec pa_play pa_recplay
+all: pa_rec pa_play pa_recplay pa_impresp
pa_rec: Makefile pa_rec.c $(SRC)
gcc $(CFLAGS) pa_rec.c $(SRC) -o pa_rec $(LIBS)
pa_recplay: Makefile pa_recplay.c $(SRC)
gcc $(CFLAGS) pa_recplay.c $(SRC) -o pa_recplay $(LIBS)
+pa_impresp: Makefile pa_impresp.c $(SRC)
+ gcc $(CFLAGS) pa_impresp.c $(SRC) -o pa_impresp $(LIBS)
+
clean:
rm -f pa_rec pa_play pa_recplay
--- /dev/null
+/*
+ pa_impresp.c
+ David Rowe
+ August 29 2012
+
+ Measures the impulse reponse of the path between the speaker and
+ microphone. Used to explore why Codec audio quality is
+ different through a speaker and headphones.
+
+ Modified from pa_playrec.c
+*/
+
+/*
+ * $Id: paex_record.c 1752 2011-09-08 03:21:55Z philburk $
+ *
+ * This program uses the PortAudio Portable Audio Library.
+ * For more information see: http://www.portaudio.com
+ * Copyright (c) 1999-2000 Ross Bencina and Phil Burk
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining
+ * a copy of this software and associated documentation files
+ * (the "Software"), to deal in the Software without restriction,
+ * including without limitation the rights to use, copy, modify, merge,
+ * publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so,
+ * subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR
+ * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
+ * CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+ * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <assert.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include "portaudio.h"
+#include "fdmdv.h"
+
+#define SAMPLE_RATE 48000 /* 48 kHz sampling rate rec. as we
+ can trust accuracy of sound
+ card */
+#define N8 160 /* processing buffer size at 8 kHz */
+#define N48 (N8*FDMDV_OS) /* processing buffer size at 48 kHz */
+#define MEM8 (FDMDV_OS_TAPS/FDMDV_OS)
+#define NUM_CHANNELS 2 /* I think most sound cards prefer
+ stereo, we will convert to mono
+ as we sample */
+
+#define IMPULSE_AMP 16384 /* amplitide of impulse */
+#define IMPULSE_PERIOD 0.1 /* period (dly between impulses) in secs */
+
+/* state information passed to call back */
+
+typedef struct {
+ float in48k[FDMDV_OS_TAPS + N48];
+ float in8k[MEM8 + N8];
+ FILE *fimp;
+ float *impulse_buf;
+ int impulse_buf_length;
+ int impulse_sample_count;
+ int framesLeft;
+} paTestData;
+
+
+/*
+ This routine will be called by the PortAudio engine when audio is
+ required. It may be called at interrupt level on some machines so
+ don't do anything that could mess up the system like calling
+ malloc() or free().
+*/
+
+static int callback( const void *inputBuffer, void *outputBuffer,
+ unsigned long framesPerBuffer,
+ const PaStreamCallbackTimeInfo* timeInfo,
+ PaStreamCallbackFlags statusFlags,
+ void *userData )
+{
+ paTestData *data = (paTestData*)userData;
+ int i;
+ short *rptr = (short*)inputBuffer;
+ short *wptr = (short*)outputBuffer;
+ float *in8k = data->in8k;
+ float *in48k = data->in48k;
+ float out8k[N8];
+ float out48k[N48];
+ short out48k_short[N48];
+ short out8k_short[N8];
+
+ (void) timeInfo;
+ (void) statusFlags;
+
+ assert(inputBuffer != NULL);
+
+ /* just use left channel */
+
+ for(i=0; i<framesPerBuffer; i++,rptr+=2)
+ data->in48k[i+FDMDV_OS_TAPS] = *rptr;
+
+ /* downsample and update filter memory */
+
+ fdmdv_48_to_8(out8k, &in48k[FDMDV_OS_TAPS], N8);
+ for(i=0; i<FDMDV_OS_TAPS; i++)
+ in48k[i] = in48k[i+framesPerBuffer];
+
+ /* write impulse response to disk */
+
+ for(i=0; i<N8; i++)
+ out8k_short[i] = out8k[i];
+ fwrite(out8k_short, sizeof(short), N8, data->fimp);
+
+ /* play side, read from impulse buffer */
+
+ for(i=0; i<N8; i++) {
+ in8k[MEM8+i] = data->impulse_buf[data->impulse_sample_count];
+ data->impulse_sample_count++;
+ if (data->impulse_sample_count == data->impulse_buf_length)
+ data->impulse_sample_count = 0;
+ }
+
+ /* upsample and update filter memory */
+
+ fdmdv_8_to_48(out48k, &in8k[MEM8], N8);
+ for(i=0; i<MEM8; i++)
+ in8k[i] = in8k[i+N8];
+
+ assert(outputBuffer != NULL);
+
+ /* write signal to both channels */
+
+ for(i=0; i<N48; i++)
+ out48k_short[i] = (short)out48k[i];
+ for(i=0; i<framesPerBuffer; i++,wptr+=2) {
+ wptr[0] = out48k_short[i];
+ wptr[1] = out48k_short[i];
+ }
+
+ data->framesLeft -= framesPerBuffer;
+ if (data->framesLeft > 0)
+ return paContinue;
+ else
+ return paComplete;
+}
+
+int main(int argc, char *argv[])
+{
+ PaStreamParameters inputParameters, outputParameters;
+ PaStream* stream;
+ PaError err = paNoError;
+ paTestData data;
+ int i, numSecs;
+
+ if (argc != 3) {
+ printf("usage: %s impulseRawFile time(s)\n", argv[0]);
+ exit(0);
+ }
+
+ data.fimp = fopen(argv[1], "wb");
+ if (data.fimp == NULL) {
+ printf("Error opening impulse output file %s\n", argv[1]);
+ exit(1);
+ }
+
+ numSecs = atoi(argv[2]);
+ data.framesLeft = numSecs * SAMPLE_RATE;
+
+ /* init filter states */
+
+ for(i=0; i<MEM8; i++)
+ data.in8k[i] = 0.0;
+ for(i=0; i<FDMDV_OS_TAPS; i++)
+ data.in48k[i] = 0.0;
+
+ /* init imupulse */
+
+ data.impulse_buf_length = IMPULSE_PERIOD*(SAMPLE_RATE/FDMDV_OS);
+ printf("%d\n",data.impulse_buf_length);
+ data.impulse_buf = (float*)malloc(data.impulse_buf_length*sizeof(float));
+ assert(data.impulse_buf != NULL);
+ data.impulse_buf[0] = IMPULSE_AMP;
+ for(i=1; i<data.impulse_buf_length; i++)
+ data.impulse_buf[i] = 0;
+ data.impulse_sample_count = 0;
+
+ err = Pa_Initialize();
+ if( err != paNoError ) goto done;
+
+ inputParameters.device = Pa_GetDefaultInputDevice(); /* default input device */
+ if (inputParameters.device == paNoDevice) {
+ fprintf(stderr,"Error: No default input device.\n");
+ goto done;
+ }
+ inputParameters.channelCount = NUM_CHANNELS; /* stereo input */
+ inputParameters.sampleFormat = paInt16;
+ inputParameters.suggestedLatency = Pa_GetDeviceInfo( inputParameters.device )->defaultLowInputLatency;
+ inputParameters.hostApiSpecificStreamInfo = NULL;
+
+ outputParameters.device = Pa_GetDefaultOutputDevice(); /* default output device */
+ if (outputParameters.device == paNoDevice) {
+ fprintf(stderr,"Error: No default output device.\n");
+ goto done;
+ }
+ outputParameters.channelCount = NUM_CHANNELS; /* stereo output */
+ outputParameters.sampleFormat = paInt16;
+ outputParameters.suggestedLatency = Pa_GetDeviceInfo( outputParameters.device )->defaultLowOutputLatency;
+ outputParameters.hostApiSpecificStreamInfo = NULL;
+
+ /* Play some audio --------------------------------------------- */
+
+ err = Pa_OpenStream(
+ &stream,
+ &inputParameters,
+ &outputParameters,
+ SAMPLE_RATE,
+ N48,
+ paClipOff,
+ callback,
+ &data );
+ if( err != paNoError ) goto done;
+
+ err = Pa_StartStream( stream );
+ if( err != paNoError ) goto done;
+
+ while( ( err = Pa_IsStreamActive( stream ) ) == 1 )
+ {
+ Pa_Sleep(100);
+ }
+ if( err < 0 ) goto done;
+
+ err = Pa_CloseStream( stream );
+ if( err != paNoError ) goto done;
+
+
+done:
+ Pa_Terminate();
+ if( err != paNoError )
+ {
+ fprintf( stderr, "An error occured while using the portaudio stream\n" );
+ fprintf( stderr, "Error number: %d\n", err );
+ fprintf( stderr, "Error message: %s\n", Pa_GetErrorText( err ) );
+ err = 1; /* Always return 0 or 1, but no other return codes. */
+ }
+
+ fclose(data.fimp);
+
+ return err;
+}
+
lsp.c \
phase.c \
quantise.c \
-ampexp.c \
pack.c \
codebook.c \
codebookd.c \
c2dec_LDADD = $(lib_LTLIBRARIES)
c2dec_LDFLAGS = $(LIBS)
-c2sim_SOURCES = c2sim.c
+c2sim_SOURCES = c2sim.c ampexp.c phaseexp.c
c2sim_LDADD = $(lib_LTLIBRARIES)
c2sim_LDFLAGS = $(LIBS)
libcodec2_la-fdmdv.lo libcodec2_la-kiss_fft.lo \
libcodec2_la-interp.lo libcodec2_la-lsp.lo \
libcodec2_la-phase.lo libcodec2_la-quantise.lo \
- libcodec2_la-ampexp.lo libcodec2_la-pack.lo \
- libcodec2_la-codebook.lo libcodec2_la-codebookd.lo \
- libcodec2_la-codebookvq.lo libcodec2_la-codebookjnd.lo \
- libcodec2_la-codebookjvm.lo libcodec2_la-codebookvqanssi.lo \
- libcodec2_la-codebookdt.lo libcodec2_la-codebookge.lo
+ libcodec2_la-pack.lo libcodec2_la-codebook.lo \
+ libcodec2_la-codebookd.lo libcodec2_la-codebookvq.lo \
+ libcodec2_la-codebookjnd.lo libcodec2_la-codebookjvm.lo \
+ libcodec2_la-codebookvqanssi.lo libcodec2_la-codebookdt.lo \
+ libcodec2_la-codebookge.lo
libcodec2_la_OBJECTS = $(am_libcodec2_la_OBJECTS)
binPROGRAMS_INSTALL = $(INSTALL_PROGRAM)
PROGRAMS = $(bin_PROGRAMS) $(noinst_PROGRAMS)
am_c2enc_OBJECTS = c2enc.$(OBJEXT)
c2enc_OBJECTS = $(am_c2enc_OBJECTS)
c2enc_DEPENDENCIES = $(am__DEPENDENCIES_1)
-am_c2sim_OBJECTS = c2sim.$(OBJEXT)
+am_c2sim_OBJECTS = c2sim.$(OBJEXT) ampexp.$(OBJEXT) phaseexp.$(OBJEXT)
c2sim_OBJECTS = $(am_c2sim_OBJECTS)
c2sim_DEPENDENCIES = $(am__DEPENDENCIES_1)
am_fdmdv_demod_OBJECTS = fdmdv_demod.$(OBJEXT) fdmdv.$(OBJEXT) \
lsp.c \
phase.c \
quantise.c \
-ampexp.c \
pack.c \
codebook.c \
codebookd.c \
c2dec_SOURCES = c2dec.c
c2dec_LDADD = $(lib_LTLIBRARIES)
c2dec_LDFLAGS = $(LIBS)
-c2sim_SOURCES = c2sim.c
+c2sim_SOURCES = c2sim.c ampexp.c phaseexp.c
c2sim_LDADD = $(lib_LTLIBRARIES)
c2sim_LDFLAGS = $(LIBS)
fdmdv_get_test_bits_SOURCES = fdmdv_get_test_bits.c fdmdv.c kiss_fft.c
distclean-compile:
-rm -f *.tab.c
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/ampexp.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/c2dec.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/c2demo.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/c2enc.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/generate_codebook.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/genlspdtcb.Po@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/kiss_fft.Po@am__quote@
-@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-ampexp.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-codebook.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-codebookd.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-codebookdt.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-quantise.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libcodec2_la-sine.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/octave.Po@am__quote@
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/phaseexp.Po@am__quote@
.c.o:
@am__fastdepCC_TRUE@ if $(COMPILE) -MT $@ -MD -MP -MF "$(DEPDIR)/$*.Tpo" -c -o $@ $<; \
@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
@am__fastdepCC_FALSE@ $(LIBTOOL) --tag=CC --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(libcodec2_la_CFLAGS) $(CFLAGS) -c -o libcodec2_la-quantise.lo `test -f 'quantise.c' || echo '$(srcdir)/'`quantise.c
-libcodec2_la-ampexp.lo: ampexp.c
-@am__fastdepCC_TRUE@ if $(LIBTOOL) --tag=CC --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(libcodec2_la_CFLAGS) $(CFLAGS) -MT libcodec2_la-ampexp.lo -MD -MP -MF "$(DEPDIR)/libcodec2_la-ampexp.Tpo" -c -o libcodec2_la-ampexp.lo `test -f 'ampexp.c' || echo '$(srcdir)/'`ampexp.c; \
-@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/libcodec2_la-ampexp.Tpo" "$(DEPDIR)/libcodec2_la-ampexp.Plo"; else rm -f "$(DEPDIR)/libcodec2_la-ampexp.Tpo"; exit 1; fi
-@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='ampexp.c' object='libcodec2_la-ampexp.lo' libtool=yes @AMDEPBACKSLASH@
-@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@
-@am__fastdepCC_FALSE@ $(LIBTOOL) --tag=CC --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(libcodec2_la_CFLAGS) $(CFLAGS) -c -o libcodec2_la-ampexp.lo `test -f 'ampexp.c' || echo '$(srcdir)/'`ampexp.c
-
libcodec2_la-pack.lo: pack.c
@am__fastdepCC_TRUE@ if $(LIBTOOL) --tag=CC --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(libcodec2_la_CFLAGS) $(CFLAGS) -MT libcodec2_la-pack.lo -MD -MP -MF "$(DEPDIR)/libcodec2_la-pack.Tpo" -c -o libcodec2_la-pack.lo `test -f 'pack.c' || echo '$(srcdir)/'`pack.c; \
@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/libcodec2_la-pack.Tpo" "$(DEPDIR)/libcodec2_la-pack.Plo"; else rm -f "$(DEPDIR)/libcodec2_la-pack.Tpo"; exit 1; fi
#include "postfilter.h"
#include "interp.h"
#include "ampexp.h"
+#include "phaseexp.h"
void synth_one_frame(kiss_fft_cfg fft_inv_cfg, short buf[], MODEL *model, float Sn_[], float Pn[]);
void print_help(const struct option *long_options, int num_opts, char* argv[]);
/* just to make sure we are not cheating - kill all phases */
- for(i=0; i<MAX_AMP; i++)
+ for(i=0; i<=MAX_AMP; i++)
model.phi[i] = 0;
if (hand_voicing) {
if (lspanssi) {
/* multi-stage VQ from Anssi Ramo OH3GDD */
- lspanssi_quantise(lsps, lsps_, LPC_ORD);
+ lspanssi_quantise(lsps, lsps_, LPC_ORD, 5);
lsp_to_lpc(lsps_, ak, LPC_ORD);
}
-10 128
- 0.3360 0.4547 0.6542 0.9336 1.1451 1.3330 1.6239 1.8289 2.1211 2.4767
- 0.2691 0.3353 0.7667 1.2615 1.4413 1.6191 1.8433 2.0450 2.5208 2.6878
- 0.2532 0.3475 0.4905 0.9081 1.4308 1.6171 1.8267 2.0767 2.3056 2.4890
- 0.1642 0.2231 0.3985 0.5937 1.0764 1.5668 2.0179 2.2026 2.5460 2.7686
- 0.1446 0.2382 0.3611 0.6294 1.4526 1.6285 1.8859 2.0916 2.5247 2.7616
- 0.2909 0.3707 0.7973 1.0222 1.2198 1.6585 1.9237 2.1546 2.5072 2.6632
- 0.2723 0.3650 0.5628 0.9811 1.2226 1.3994 1.9010 2.1329 2.3930 2.5913
- 0.3328 0.4212 0.6131 0.9549 1.4766 1.6833 1.8794 2.1090 2.3587 2.5268
- 0.1916 0.3023 0.4474 0.7898 1.3276 1.4953 1.7983 2.0377 2.2989 2.7252
- 0.2257 0.3252 0.8949 1.1911 1.4134 1.6653 1.9306 2.1786 2.4877 2.6719
- 0.1705 0.2897 0.4621 0.7049 0.8953 1.1664 1.5965 1.8166 2.5390 2.7622
- 0.2085 0.3298 0.5015 0.7248 0.8937 1.3994 1.8019 2.0194 2.3466 2.5436
- 0.1516 0.2919 0.9066 1.2688 1.5725 1.8833 2.1739 2.3893 2.6362 2.8002
- 0.1590 0.2264 0.4001 0.5947 0.9463 1.5007 1.8331 2.0431 2.2922 2.5175
- 0.2365 0.3283 0.8137 1.0816 1.2930 1.5331 1.8063 2.0186 2.2931 2.5537
- 0.2813 0.5931 1.0526 1.4005 1.6873 1.9747 2.2304 2.4244 2.6560 2.8101
- 0.2161 0.3113 0.4777 0.6719 0.8755 1.7048 1.9748 2.2109 2.5294 2.7116
- 0.2145 0.2934 0.4785 0.8062 1.6183 1.8578 2.0875 2.2896 2.5415 2.7191
- 0.3137 0.3927 0.6740 0.8238 1.0395 1.7024 1.9657 2.1882 2.5408 2.6864
- 0.2782 0.4392 0.6951 0.9501 1.3151 1.5618 1.8651 2.1550 2.4891 2.7237
- 0.2801 0.3536 0.7369 1.1399 1.3120 1.5235 1.7564 1.9534 2.5079 2.6729
- 0.1942 0.2861 0.4241 0.9135 1.4995 1.6658 1.8824 2.1044 2.3794 2.6924
- 0.1848 0.2642 0.4669 0.8099 1.1814 1.3632 1.8145 2.0406 2.3029 2.4852
- 0.1876 0.2791 0.4590 0.7803 0.9716 1.2344 1.7653 1.9954 2.3031 2.4919
- 0.2118 0.3546 0.5173 0.8545 1.0944 1.2773 1.5391 1.7636 2.5227 2.7473
- 0.3702 0.5754 0.7567 0.9470 1.2418 1.4175 1.6937 1.9781 2.2335 2.4379
- 0.2994 0.3896 0.7103 0.9276 1.1109 1.5065 1.7855 1.9749 2.5348 2.6989
- 0.1669 0.2410 0.4275 0.6556 1.0301 1.2090 1.6554 1.9960 2.2732 2.7085
- 0.2147 0.3188 0.4951 1.1641 1.3285 1.5939 1.8097 2.2924 2.5835 2.7804
- 0.2993 0.4572 0.6620 0.8856 1.0647 1.3143 1.9459 2.1805 2.4665 2.6892
- 0.2834 0.4582 1.1467 1.4071 1.6675 1.8805 2.0877 2.2818 2.5478 2.7149
- 0.4616 0.6943 1.0694 1.3309 1.6226 1.8894 2.2023 2.3914 2.6245 2.7663
- 0.3407 0.4436 0.7198 0.9888 1.1796 1.3665 1.5898 1.8618 2.4254 2.5902
- 0.2732 0.3898 0.5375 0.8858 1.2724 1.4400 1.7338 1.9968 2.2495 2.4836
- 0.2131 0.3241 0.4965 0.7160 1.2157 1.3966 1.6804 2.1688 2.4174 2.6745
- 0.2469 0.4291 0.6036 0.7812 1.2001 1.3969 1.6161 2.0469 2.3207 2.5517
- 0.1485 0.2434 0.4256 1.0104 1.3060 1.5364 1.9441 2.1716 2.5517 2.7629
- 0.1874 0.2675 0.4621 0.6966 1.2937 1.4904 1.7493 1.9962 2.2875 2.4444
- 0.3439 0.5125 0.8961 1.2283 1.6415 1.8526 2.1285 2.3249 2.5596 2.7308
- 0.1505 0.2717 0.7294 0.9670 1.2517 1.4743 1.8532 2.1339 2.4972 2.7315
- 0.1350 0.1930 0.3427 0.5386 0.8454 1.1797 1.6950 1.9516 2.3599 2.7012
- 0.2176 0.3366 0.5235 0.9077 1.0848 1.5432 1.8776 2.1216 2.5428 2.7217
- 0.3014 0.3904 0.6168 0.7906 0.9655 1.5106 1.7991 1.9911 2.5273 2.6787
- 0.1367 0.2172 0.3931 0.8468 1.1326 1.3594 1.6932 1.8975 2.3663 2.6332
- 0.2887 0.4712 0.6418 0.8754 1.3482 1.5325 1.7649 2.0353 2.3070 2.4830
- 0.1483 0.2244 0.6148 0.8199 1.0681 1.3115 1.7512 2.0202 2.4751 2.7189
- 0.4893 0.7042 0.9602 1.1855 1.4384 1.6609 1.9721 2.2007 2.4766 2.6608
- 0.2090 0.2716 0.5730 0.8007 1.4094 1.6658 1.9273 2.2034 2.5187 2.7289
- 0.2523 0.3927 0.5506 0.7813 1.3441 1.5397 1.7531 2.1277 2.3804 2.5818
- 0.1522 0.2343 0.4990 1.0387 1.4201 1.7548 2.1013 2.3461 2.6402 2.8172
- 0.1769 0.2704 0.4287 0.7267 1.2940 1.4642 1.9797 2.2008 2.4983 2.7310
- 0.3926 0.5534 0.9519 1.1643 1.3749 1.5545 1.8195 2.0650 2.4053 2.6192
- 0.1216 0.1753 0.3177 0.5209 1.1123 1.3918 1.7803 2.0468 2.4208 2.7579
- 0.2275 0.4219 0.5920 0.7925 1.0098 1.1834 1.6747 1.9399 2.2123 2.6777
- 0.3513 0.4333 0.7414 1.1633 1.3557 1.5613 1.9002 2.1023 2.4813 2.6792
- 0.1805 0.2762 0.4687 1.0362 1.2487 1.4739 1.7645 1.9964 2.3579 2.5870
- 0.1650 0.2325 0.3927 0.5699 1.3112 1.5287 1.8092 2.0778 2.3406 2.6345
- 0.2073 0.3261 0.5129 0.7185 1.1010 1.2870 1.5341 2.0697 2.3255 2.5546
- 0.2637 0.3758 0.5427 1.0661 1.2785 1.4782 1.8293 2.0197 2.5442 2.7447
- 0.3518 0.4465 0.6788 1.0214 1.2058 1.3951 1.8217 2.0068 2.2953 2.6047
- 0.2520 0.3495 0.5606 1.0545 1.4043 1.5693 1.9423 2.1639 2.4292 2.6359
- 0.2052 0.3142 0.4657 0.6797 0.8644 1.0609 1.6940 1.9925 2.3114 2.5765
- 0.1597 0.2850 0.4431 0.8797 1.1379 1.3253 1.8077 1.9974 2.5697 2.7600
- 0.2704 0.5130 0.7074 0.9081 1.2218 1.3973 1.7209 2.0897 2.3723 2.5672
- 0.1979 0.3157 0.4424 0.6863 1.3914 1.6539 1.8489 2.1366 2.3713 2.5897
- 0.1831 0.2738 0.4569 0.7913 1.1047 1.3384 1.6874 1.9033 2.1215 2.2992
- 0.1547 0.2436 0.4290 0.6646 0.8508 1.4500 1.7414 2.0040 2.5020 2.7217
- 0.2365 0.3103 0.6833 1.0242 1.6000 1.8179 2.0437 2.2433 2.4953 2.6656
- 0.3520 0.4860 0.8339 1.1147 1.5155 1.7047 1.9530 2.1948 2.4500 2.6365
- 0.2076 0.2908 0.7235 0.9151 1.1783 1.3678 1.7172 2.0573 2.3534 2.5666
- 0.1728 0.2792 0.4252 0.8989 1.3595 1.5128 1.7950 1.9827 2.5640 2.7569
- 0.1745 0.2588 0.4083 0.5967 1.1944 1.7594 1.9682 2.1809 2.4264 2.6352
- 0.1576 0.2379 0.3844 0.9002 1.5739 1.7294 1.9991 2.2252 2.5406 2.7395
- 0.1590 0.2374 0.4064 0.6555 1.2164 1.4072 1.7393 1.9394 2.5174 2.7430
- 0.4886 0.6711 0.8582 1.0439 1.2552 1.4417 1.8238 2.0966 2.3791 2.5812
- 0.2051 0.3085 0.4398 0.7621 1.4911 1.8260 1.9903 2.1831 2.4109 2.5801
- 0.3120 0.4273 0.6226 0.8492 1.0291 1.2194 1.7417 2.0374 2.2845 2.4926
- 0.2005 0.3242 0.4832 0.7931 1.0221 1.2054 1.8914 2.1318 2.4332 2.7511
- 0.2252 0.4091 0.5698 0.8823 1.1331 1.3113 1.7990 1.9996 2.3674 2.7118
- 0.4395 0.6887 1.1368 1.3646 1.5815 1.7778 2.0206 2.2271 2.5163 2.6897
- 0.3068 0.3953 0.6710 0.8584 1.0526 1.5481 1.8187 2.0131 2.3237 2.5077
- 0.3999 0.5240 0.7133 0.9845 1.3603 1.5261 1.7888 2.0948 2.3560 2.5175
- 0.4128 0.5257 0.7499 0.9281 1.1052 1.5338 1.8865 2.0833 2.5113 2.6860
- 0.2268 0.5364 0.8215 1.0488 1.2645 1.4868 1.8685 2.1692 2.4996 2.7297
- 0.2220 0.3369 0.4770 0.6673 0.8402 1.0480 1.9389 2.2364 2.5056 2.7767
- 0.4037 0.5540 0.7367 0.9263 1.1114 1.2888 1.7503 2.0741 2.3214 2.5323
- 0.2521 0.3693 0.5398 0.7362 0.9155 1.0727 1.5763 2.1858 2.4527 2.6618
- 0.1489 0.2149 0.3727 0.5700 0.9049 1.1900 1.6205 1.8790 2.1601 2.3320
- 0.1847 0.2632 0.4350 1.3060 1.5308 1.7242 2.0072 2.2530 2.5575 2.7530
- 0.2075 0.3405 0.5168 0.8045 1.1630 1.3270 1.6556 1.8807 2.1707 2.7007
- 0.2232 0.3092 0.4996 0.7629 1.2442 1.8194 2.0729 2.2511 2.5128 2.6962
- 0.2044 0.2831 0.6379 0.8276 1.2594 1.5083 1.7621 1.9768 2.3941 2.5925
- 0.2566 0.4715 0.6754 0.8465 1.0929 1.2802 1.5230 2.0233 2.3058 2.5254
- 0.1728 0.2480 0.3956 0.5761 1.3228 1.8431 2.0752 2.2605 2.5850 2.7719
- 0.2336 0.3500 0.5382 0.7161 1.1749 1.6151 1.8457 2.1034 2.5132 2.6961
- 0.1626 0.2245 0.4010 0.5814 1.1522 1.3662 1.6484 1.9770 2.2715 2.4647
- 0.5064 0.8353 1.2534 1.4713 1.7581 1.9661 2.1943 2.3698 2.5851 2.7201
- 0.1130 0.2084 0.7565 1.0949 1.4016 1.6644 1.9825 2.2510 2.5828 2.7910
- 0.2126 0.2938 0.7026 0.9381 1.4298 1.6365 1.8854 2.1055 2.4127 2.6110
- 0.1614 0.2645 0.4155 0.6642 0.8507 1.1249 1.8711 2.0782 2.5291 2.7432
- 0.2586 0.3414 0.4880 1.0008 1.5909 1.8256 2.0031 2.1976 2.4261 2.5968
- 0.1673 0.2558 0.3886 0.5882 1.5263 1.7804 1.9941 2.2378 2.4797 2.7040
- 0.1448 0.2216 0.4007 0.7222 1.0049 1.4014 1.7176 1.9074 2.1345 2.6733
- 0.2047 0.4294 0.5915 0.8014 0.9788 1.1973 1.7570 1.9584 2.5234 2.7386
- 0.2475 0.3425 0.5105 0.7038 0.8549 1.3804 1.9609 2.1499 2.5170 2.6911
- 0.1982 0.2876 0.4529 1.1780 1.4156 1.5881 1.8443 2.0405 2.5394 2.7164
- 0.2335 0.4039 0.7689 1.1114 1.3960 1.7002 2.0436 2.3134 2.6207 2.8013
- 0.1507 0.2238 0.4358 0.7731 1.0890 1.5243 1.8333 2.0400 2.3737 2.7067
- 0.2541 0.3993 0.6220 0.8230 1.1039 1.2991 1.5801 2.1739 2.5225 2.7008
- 0.2543 0.3728 0.5599 0.8109 1.4458 1.7316 1.9325 2.1289 2.3867 2.5548
- 0.2260 0.3130 0.6481 0.8326 1.0378 1.3403 1.6607 1.9169 2.2469 2.4594
- 0.3529 0.5530 0.8657 1.0988 1.3804 1.6333 1.9592 2.2094 2.5405 2.7394
- 0.1026 0.1545 0.3283 0.8287 1.2123 1.5327 1.8629 2.1381 2.5010 2.7817
- 0.3150 0.5394 0.9199 1.2263 1.4777 1.8015 2.1354 2.3972 2.6672 2.8213
- 0.1836 0.2766 0.4200 0.5903 0.7749 0.9537 1.6318 2.1226 2.4240 2.7110
- 0.2997 0.3708 0.6137 1.1349 1.3715 1.5328 1.8146 2.0103 2.2614 2.5400
- 0.4199 0.5157 0.7955 1.0556 1.2265 1.4824 1.8047 2.0097 2.4818 2.6784
- 0.2618 0.3266 0.5846 1.1892 1.5473 1.6985 1.9038 2.1166 2.3609 2.5916
- 0.0852 0.1507 0.5832 0.9281 1.2390 1.5409 1.8268 2.1237 2.4848 2.7556
- 0.2296 0.3152 0.6005 0.9924 1.2153 1.4437 1.7335 1.9371 2.1647 2.3545
- 0.1950 0.2658 0.5889 1.2970 1.6235 1.7986 2.1255 2.3258 2.5837 2.7473
- 0.2120 0.3353 0.5072 0.7848 0.9818 1.1919 1.4731 1.7419 2.1646 2.4352
- 0.3148 0.4107 0.5890 0.7912 0.9374 1.1670 1.8583 2.1654 2.4073 2.6118
- 0.3546 0.5179 1.0110 1.2769 1.4836 1.6974 1.9885 2.2092 2.5144 2.7007
- 0.2342 0.3076 0.8029 1.3718 1.5773 1.7549 2.0162 2.2209 2.5384 2.7040
- 0.3588 0.4791 0.6896 0.8763 1.0393 1.2816 1.7453 1.9395 2.4555 2.6853
- 0.2030 0.2859 0.5600 0.7655 1.0689 1.5931 1.8341 2.0201 2.2169 2.4952
- 0.2469 0.3202 0.6416 1.2561 1.4636 1.6476 1.9578 2.1515 2.5350 2.7118
+10 256
+ 0.1758 0.2325 0.3916 0.5606 1.0626 1.5769 2.1164 2.2761 2.5949 2.7999
+ 0.1521 0.2404 0.3772 0.9015 1.4054 1.5579 1.7947 1.9811 2.4882 2.6900
+ 0.4223 0.6056 0.9696 1.1914 1.3841 1.5672 1.8329 2.0743 2.4116 2.6234
+ 0.1313 0.2163 0.4031 1.0425 1.3249 1.5738 1.9110 2.1936 2.5496 2.7544
+ 0.2999 0.3797 0.5583 0.9155 1.5153 1.8028 1.9851 2.1696 2.4251 2.5761
+ 0.2125 0.3332 0.4977 0.7872 0.9833 1.1857 1.4595 1.7210 2.1291 2.3781
+ 0.1591 0.2184 0.3912 0.5746 0.8492 1.4360 1.7276 1.9613 2.2853 2.5012
+ 0.2197 0.2949 0.6634 0.8397 1.0910 1.3576 1.6756 1.8970 2.3438 2.5339
+ 0.4189 0.5330 0.7211 1.0209 1.3794 1.5377 1.7934 2.0949 2.3568 2.5091
+ 0.1257 0.1993 0.6106 1.0936 1.4385 1.6787 1.9458 2.1916 2.5018 2.7395
+ 0.2827 0.4198 0.5761 0.8855 1.2018 1.3521 1.5792 1.7611 2.0888 2.6451
+ 0.1798 0.3052 0.4330 0.7897 1.3711 1.5213 1.8613 2.0291 2.5364 2.7726
+ 0.0883 0.1298 0.3197 0.8644 1.2367 1.5544 1.8702 2.1623 2.5103 2.7934
+ 0.1668 0.2484 0.4062 0.7731 1.4153 1.6018 1.9432 2.1927 2.4356 2.6769
+ 0.2660 0.4510 0.6057 0.7803 1.2113 1.4123 1.6018 2.0197 2.2951 2.5269
+ 0.1970 0.2923 0.4497 0.6660 1.2242 1.7391 1.9464 2.1688 2.4235 2.6251
+ 0.2043 0.3011 0.4879 1.2240 1.3719 1.6341 1.8456 2.3066 2.5951 2.7749
+ 0.1960 0.3449 0.5090 0.8307 1.0628 1.2440 1.5223 1.7436 2.5741 2.7868
+ 0.3422 0.4983 1.0662 1.2989 1.4978 1.6881 1.9339 2.1477 2.4571 2.6527
+ 0.1725 0.2411 0.4273 0.6199 1.0212 1.2024 1.5354 2.0466 2.3228 2.6426
+ 0.1752 0.2453 0.3888 0.5690 1.3755 1.8733 2.0960 2.2699 2.5990 2.7772
+ 0.2163 0.2928 0.5757 1.1500 1.4922 1.6758 2.0768 2.2616 2.5238 2.7015
+ 0.1527 0.2148 0.3752 0.5928 1.1300 1.3139 1.7715 1.9751 2.2517 2.7336
+ 0.2793 0.3756 0.5507 0.7341 0.8864 1.4654 1.9608 2.1310 2.4776 2.6431
+ 0.1568 0.2467 0.3955 0.8493 1.2988 1.4984 2.0480 2.2761 2.6066 2.8025
+ 0.2211 0.3091 0.6485 1.1112 1.3016 1.5266 1.8555 2.0720 2.3781 2.5747
+ 0.2012 0.2778 0.7755 1.3916 1.6557 1.8476 2.1327 2.3197 2.5862 2.7458
+ 0.4200 0.6053 0.8969 1.1622 1.4849 1.6800 1.9658 2.1983 2.4634 2.6417
+ 0.1524 0.2137 0.3660 0.5431 1.0506 1.5835 1.8941 2.0930 2.3506 2.6062
+ 0.2539 0.3298 0.4880 1.0866 1.6389 1.8590 2.0297 2.2238 2.4305 2.5985
+ 0.1490 0.2354 0.3523 0.5899 1.5254 1.7098 1.9930 2.2424 2.5434 2.7675
+ 0.3292 0.5926 0.7600 0.9593 1.3205 1.4874 1.7354 2.0276 2.2777 2.4770
+ 0.2279 0.3568 0.5061 0.9325 1.1228 1.3460 1.9175 2.1015 2.5629 2.7649
+ 0.3251 0.4002 0.6573 1.1360 1.3333 1.4974 1.8122 1.9791 2.3106 2.6294
+ 0.1715 0.2795 0.4261 1.0318 1.2758 1.4591 1.8327 2.0086 2.6243 2.7997
+ 0.2652 0.3435 0.4778 0.9549 1.5217 1.7191 1.8950 2.1142 2.3158 2.4999
+ 0.3935 0.5384 0.7142 0.8996 1.2181 1.3894 1.6082 1.9749 2.3746 2.5363
+ 0.2416 0.3049 0.8079 1.3759 1.5411 1.7090 1.9377 2.1433 2.5408 2.6971
+ 0.2059 0.3193 0.4920 0.8434 1.0095 1.4318 1.7810 1.9819 2.4335 2.6197
+ 0.4156 0.5149 0.7530 0.9292 1.1012 1.6025 1.8856 2.0881 2.5362 2.6869
+ 0.2269 0.4375 0.5934 0.7849 0.9785 1.1460 1.6544 1.8826 2.1561 2.6623
+ 0.3085 0.3934 0.6059 0.7601 0.9255 1.3825 1.7690 1.9514 2.3718 2.5482
+ 0.1896 0.3181 0.4753 0.7925 0.9724 1.1929 1.9433 2.1486 2.5434 2.7600
+ 0.1163 0.1776 0.3435 0.7213 1.0756 1.3618 1.8124 2.0266 2.4264 2.7561
+ 0.1640 0.2611 0.4637 0.6976 0.8929 1.3606 1.6637 1.8628 2.5272 2.7552
+ 0.1921 0.2518 0.5555 0.7407 1.3333 1.5577 1.7845 2.1064 2.4157 2.6185
+ 0.2331 0.3536 0.5005 0.7809 1.3358 1.5083 1.7333 2.1940 2.4437 2.6632
+ 0.2056 0.2853 0.6215 0.8327 1.1414 1.5483 1.7742 1.9407 2.1665 2.5996
+ 0.3208 0.4780 0.9442 1.3425 1.5282 1.7582 2.0464 2.2526 2.5865 2.7716
+ 0.2752 0.3570 0.6803 1.0739 1.2423 1.6180 1.8955 2.0892 2.5536 2.7164
+ 0.1652 0.2401 0.3748 0.5366 1.1697 1.8183 2.0168 2.2192 2.4511 2.6629
+ 0.1923 0.2578 0.5457 0.7913 1.0920 1.3526 1.8263 2.0704 2.3184 2.4883
+ 0.2224 0.3361 0.5039 0.9147 1.1599 1.3169 1.8463 2.0467 2.3014 2.6788
+ 0.2708 0.4679 0.6315 0.8213 0.9755 1.2014 1.8963 2.1636 2.4957 2.7522
+ 0.2313 0.3366 0.4789 0.6823 0.8660 1.0114 1.6931 2.3064 2.5454 2.7355
+ 0.3929 0.4783 0.8189 1.0752 1.2229 1.5766 1.9033 2.0991 2.5274 2.6876
+ 0.1614 0.4036 0.5796 0.8109 1.0455 1.2314 1.8061 1.9904 2.4918 2.7303
+ 0.2238 0.3274 0.4851 1.0162 1.2151 1.4275 1.6030 2.2531 2.5822 2.7636
+ 0.1496 0.2156 0.5937 0.7803 1.0021 1.2467 1.7271 2.0531 2.4477 2.7164
+ 0.2352 0.3707 0.7219 1.0819 1.4089 1.7501 2.0830 2.3376 2.6257 2.7939
+ 0.4415 0.5969 0.7706 0.9507 1.1369 1.3044 1.7731 2.1115 2.3365 2.5230
+ 0.2695 0.5174 0.7114 0.9150 1.2299 1.3929 1.7219 2.1080 2.3765 2.5648
+ 0.3723 0.5722 1.2061 1.4462 1.7032 1.9061 2.1227 2.3038 2.5600 2.7230
+ 0.1831 0.2880 0.4092 0.6519 1.4638 1.7207 1.9117 2.1251 2.3508 2.5557
+ 0.1102 0.2082 0.7891 1.1167 1.3907 1.6560 1.9993 2.2804 2.6209 2.8167
+ 0.2266 0.3315 0.5169 0.9196 1.0896 1.6117 1.9061 2.1366 2.5239 2.7008
+ 0.1925 0.2798 0.4606 0.7672 1.3114 1.4917 1.7614 1.9867 2.2726 2.4307
+ 0.2374 0.3578 0.5345 0.7479 1.0785 1.2451 1.6256 2.1214 2.3719 2.6338
+ 0.3256 0.4565 0.8271 1.1259 1.5653 1.7512 2.0054 2.2188 2.4539 2.6376
+ 0.2641 0.3886 0.5392 0.8270 1.2268 1.3692 1.7609 2.0394 2.2798 2.4842
+ 0.2292 0.3151 0.5033 0.6891 1.2249 1.7483 1.9852 2.2207 2.6241 2.7848
+ 0.1589 0.2401 0.4610 0.7210 0.9040 1.4414 1.8311 2.0575 2.3484 2.6672
+ 0.1524 0.2251 0.3907 0.6566 0.9320 1.1413 1.8153 2.0413 2.5131 2.7433
+ 0.0548 0.1327 0.6153 0.8976 1.2395 1.5467 1.8695 2.1401 2.5115 2.7860
+ 0.1795 0.2856 0.6961 0.8736 1.0785 1.3317 1.8714 2.1495 2.4956 2.7196
+ 0.2300 0.3053 0.4911 0.8705 1.6588 1.8946 2.1084 2.3044 2.5459 2.7079
+ 0.2635 0.3343 0.7564 1.1663 1.3193 1.5151 1.7293 1.9271 2.5175 2.6794
+ 0.2207 0.3314 0.4602 0.8075 1.4348 1.6423 1.8436 2.0873 2.3053 2.4962
+ 0.1585 0.2374 0.3722 0.8613 1.5772 1.7433 2.0593 2.2968 2.5824 2.7682
+ 0.2304 0.4589 0.6449 0.8472 1.1364 1.3054 1.7034 1.9975 2.2511 2.7122
+ 0.2166 0.3117 0.5057 0.6720 1.0672 1.5733 1.7716 2.1047 2.3765 2.5770
+ 0.2986 0.4949 0.6756 0.8920 1.1973 1.3641 1.7307 1.9937 2.2331 2.4665
+ 0.1523 0.2185 0.4022 0.8087 1.2387 1.4784 1.8394 2.0542 2.3875 2.6580
+ 0.1720 0.2535 0.3862 0.5572 1.4541 1.7965 1.9742 2.1872 2.4175 2.6449
+ 0.1745 0.2513 0.3982 0.8738 1.5685 1.7377 1.9246 2.1456 2.4042 2.7151
+ 0.1912 0.2525 0.6574 0.8539 1.2780 1.4585 1.7809 2.0333 2.3286 2.5247
+ 0.2338 0.3282 0.5605 1.0661 1.3060 1.5106 2.0005 2.2183 2.5211 2.7188
+ 0.1804 0.2678 0.4272 0.6533 0.8381 1.0226 1.7006 1.9531 2.3067 2.5542
+ 0.5629 0.9651 1.2818 1.4798 1.7544 1.9414 2.1655 2.3402 2.5563 2.6872
+ 0.2084 0.3219 0.4788 0.7050 0.8887 1.0577 1.7453 2.0386 2.2877 2.7504
+ 0.2365 0.6242 0.9315 1.1573 1.3936 1.6484 1.9768 2.2345 2.5289 2.7227
+ 0.3701 0.4643 0.7065 1.0603 1.2288 1.4150 1.9021 2.0919 2.3629 2.5844
+ 0.3189 0.4096 0.6377 0.8815 1.0448 1.2501 1.4944 1.6774 2.3905 2.6020
+ 0.1522 0.2386 0.3880 0.6325 1.2665 1.4377 1.7310 1.9141 2.5610 2.7797
+ 0.3117 0.3923 0.5812 0.7486 0.8815 1.2034 1.9645 2.1964 2.4322 2.6216
+ 0.2827 0.4232 0.5786 0.8290 1.4322 1.6840 1.8718 2.0862 2.3631 2.5200
+ 0.3438 0.4998 0.8579 1.1283 1.5329 1.8031 2.1248 2.3573 2.5990 2.7643
+ 0.1212 0.1709 0.3039 0.5338 1.2153 1.4955 1.8535 2.1107 2.4890 2.7774
+ 0.2421 0.3119 0.7298 1.0887 1.6384 1.8612 2.0820 2.2694 2.4988 2.6617
+ 0.2095 0.2764 0.5724 0.8658 1.4650 1.6787 1.9541 2.2196 2.5420 2.7509
+ 0.2191 0.3270 0.4370 0.8278 1.5300 1.8285 1.9944 2.1749 2.4027 2.5602
+ 0.2094 0.3048 0.4604 0.6568 0.8430 1.7034 1.9470 2.1657 2.4619 2.6636
+ 0.2469 0.3806 0.5348 0.8046 0.9852 1.1631 1.8391 2.0935 2.3307 2.7203
+ 0.2496 0.4449 0.6090 0.7998 0.9565 1.1630 1.6812 1.8863 2.5161 2.7114
+ 0.1438 0.2054 0.3547 0.5545 0.8644 1.1426 1.5377 1.8044 2.1109 2.2725
+ 0.2908 0.3648 0.7277 0.9116 1.0909 1.5579 1.7850 1.9939 2.5543 2.6963
+ 0.2133 0.2985 0.6842 0.8861 1.3707 1.6107 1.8178 1.9834 2.4537 2.6553
+ 0.2895 0.4700 0.6332 0.8526 1.3345 1.5188 1.7413 2.0823 2.4063 2.5756
+ 0.1578 0.2169 0.3704 0.5288 0.9807 1.2198 1.7834 2.0368 2.2934 2.4844
+ 0.2194 0.5249 0.8201 1.0401 1.2425 1.4544 1.8444 2.1393 2.4807 2.7174
+ 0.3707 0.4796 0.6803 0.9829 1.1510 1.3575 1.7494 1.9194 2.4520 2.6604
+ 0.2347 0.3743 0.5113 0.8843 1.2936 1.4434 1.7405 1.9574 2.2134 2.6639
+ 0.1897 0.2704 0.4872 1.0280 1.2828 1.4817 1.7984 1.9971 2.2526 2.4473
+ 0.3061 0.4827 0.7275 0.8994 1.0813 1.2700 1.6292 2.0770 2.5188 2.6817
+ 0.1806 0.2611 0.4129 0.6188 1.2706 1.4448 1.9339 2.2086 2.4997 2.7195
+ 0.2047 0.3245 0.5019 0.7067 1.1818 1.3708 1.5811 2.0002 2.2534 2.4636
+ 0.3587 0.5011 0.8490 1.0702 1.3980 1.6004 1.8911 2.1472 2.4435 2.6474
+ 0.2596 0.3730 0.5779 0.7402 1.2139 1.6142 1.8509 2.0492 2.5275 2.7046
+ 0.2271 0.3349 0.4802 1.0031 1.3797 1.5277 1.8661 2.0556 2.4635 2.7341
+ 0.1396 0.2158 0.5107 1.0619 1.3869 1.7494 2.1243 2.3798 2.6589 2.8261
+ 0.3684 0.4655 0.6615 0.8541 1.0034 1.1941 1.8027 2.1481 2.3677 2.5518
+ 0.2088 0.3135 0.9364 1.2445 1.4724 1.7126 1.9704 2.2088 2.4965 2.6824
+ 0.1733 0.2577 0.4400 0.8798 1.1278 1.3281 1.7297 1.9569 2.1910 2.5702
+ 0.3565 0.5674 0.7572 0.9267 1.1044 1.3307 1.8773 2.1678 2.4682 2.7092
+ 0.1928 0.2750 0.5010 0.7396 0.9972 1.5784 1.8464 2.0140 2.2003 2.4308
+ 0.3765 0.4597 0.7343 1.0932 1.2558 1.4780 1.7493 1.9366 2.5092 2.6906
+ 0.1800 0.2482 0.4743 1.2136 1.6519 1.8203 2.1425 2.3563 2.6016 2.7602
+ 0.2668 0.3960 0.5822 0.8525 1.0589 1.2445 1.7485 1.9852 2.2436 2.4535
+ 0.2348 0.3653 0.4907 0.7280 1.4211 1.7185 1.8921 2.1629 2.4253 2.6101
+ 0.1881 0.2837 0.7011 0.8638 1.2560 1.4354 1.7457 2.1517 2.4840 2.6841
+ 0.1726 0.2317 0.3971 0.5719 1.2753 1.4877 1.8210 2.0346 2.2805 2.4501
+ 0.2021 0.2694 0.4830 0.7346 1.1601 1.8231 2.1155 2.2877 2.5468 2.7307
+ 0.2253 0.3033 0.5790 0.7854 1.2533 1.7287 1.9596 2.1398 2.3553 2.5466
+ 0.1606 0.2248 0.3847 0.5574 1.1029 1.3482 1.5998 1.9156 2.1986 2.3804
+ 0.4353 0.6414 1.0148 1.2530 1.4888 1.7403 2.0825 2.3003 2.5836 2.7395
+ 0.1784 0.2818 0.4341 0.7948 0.9689 1.2203 1.7275 1.9251 2.4207 2.6003
+ 0.1762 0.2560 0.4179 0.7303 0.9629 1.2267 1.7455 1.9819 2.2271 2.4031
+ 0.3984 0.5801 0.9777 1.2731 1.6563 1.8474 2.0826 2.2676 2.4987 2.6701
+ 0.1354 0.2314 0.7761 1.0489 1.2961 1.5164 1.8681 2.1520 2.4995 2.7470
+ 0.2991 0.3988 0.5915 1.0276 1.2258 1.4250 1.8799 2.0505 2.4493 2.6856
+ 0.2010 0.3027 0.4291 0.8210 1.4090 1.5818 1.8218 2.0340 2.2635 2.7229
+ 0.2308 0.2929 0.6434 1.3581 1.5807 1.7420 2.0403 2.2290 2.5418 2.7080
+ 0.2994 0.3718 0.7711 0.9572 1.1657 1.7457 1.9881 2.1990 2.5254 2.6661
+ 0.2109 0.3561 0.6545 0.8223 1.1384 1.3517 1.5612 2.0152 2.4674 2.6558
+ 0.2393 0.3220 0.4971 1.0610 1.5257 1.6839 1.9273 2.1383 2.4156 2.6338
+ 0.1683 0.2504 0.3877 0.5873 1.4056 1.6155 1.8563 2.0790 2.3342 2.7347
+ 0.3831 0.5037 0.7021 0.8808 1.0336 1.2668 1.7299 1.9225 2.3112 2.6202
+ 0.2773 0.3402 0.7620 1.2645 1.4406 1.6097 1.8339 2.0245 2.5119 2.6789
+ 0.3363 0.5632 0.9721 1.2637 1.4991 1.8532 2.1985 2.4537 2.6995 2.8389
+ 0.3280 0.5088 0.8570 1.1350 1.3925 1.6439 1.9898 2.2625 2.6084 2.8047
+ 0.1597 0.2245 0.5593 0.9273 1.2507 1.5549 1.8541 2.1311 2.4626 2.7196
+ 0.3347 0.4067 0.6692 0.8024 1.0221 1.7071 1.9284 2.1468 2.5107 2.6445
+ 0.2847 0.3728 0.5354 0.9063 1.3961 1.5670 1.7828 2.0433 2.2688 2.4525
+ 0.2610 0.3508 0.8032 0.9859 1.2244 1.5762 1.8581 2.1030 2.4214 2.6103
+ 0.1752 0.3509 0.4991 0.7944 1.2337 1.3808 1.7722 1.9542 2.4782 2.7242
+ 0.1469 0.2596 0.4097 0.8394 1.1407 1.3077 1.8196 2.0045 2.5870 2.7816
+ 0.2211 0.3087 0.5999 0.8710 1.1798 1.4936 1.7728 1.9554 2.1406 2.3143
+ 0.1509 0.2284 0.4038 0.7064 1.0916 1.5342 1.8030 1.9716 2.2134 2.6886
+ 0.1876 0.4389 0.7433 0.9568 1.2191 1.5924 1.9645 2.2533 2.5997 2.7830
+ 0.2436 0.4034 0.5910 0.7788 1.2218 1.3931 1.6745 2.1690 2.4051 2.6372
+ 0.1067 0.1707 0.4182 0.9338 1.1840 1.4622 1.7338 1.9712 2.3144 2.6564
+ 0.1924 0.2911 0.4683 0.6795 1.1941 1.3722 1.5996 2.1404 2.3909 2.6000
+ 0.3157 0.4695 0.6310 0.8589 1.3236 1.5054 1.7412 2.0118 2.2522 2.4449
+ 0.3348 0.4388 0.6798 0.8653 1.0403 1.4575 2.0259 2.2035 2.5229 2.6958
+ 0.1513 0.4688 0.8822 1.2424 1.4726 1.7809 2.1290 2.3844 2.6710 2.8350
+ 0.1650 0.2578 0.4115 0.8528 1.1635 1.3411 1.5985 1.7965 2.4082 2.6618
+ 0.1271 0.2409 0.8401 1.2151 1.5597 1.8711 2.1836 2.4189 2.6611 2.8231
+ 0.1496 0.4338 0.7746 1.1129 1.4141 1.6232 1.9263 2.2016 2.5130 2.7624
+ 0.1905 0.2838 0.4213 0.6682 1.5214 1.8759 2.0589 2.2379 2.4936 2.6636
+ 0.2456 0.4494 0.5896 0.9095 1.1140 1.2965 1.7243 1.9015 2.4002 2.6808
+ 0.2169 0.3407 0.5440 0.7308 0.9420 1.1116 1.4378 2.0193 2.3458 2.6192
+ 0.2677 0.4433 0.6157 0.7994 1.1118 1.5315 1.7649 2.1983 2.6092 2.7728
+ 0.1943 0.3031 0.4438 0.6802 1.0895 1.2501 1.8642 2.1387 2.4124 2.7289
+ 0.4297 0.7171 1.1938 1.4019 1.5945 1.7721 2.0149 2.2262 2.5182 2.6865
+ 0.2817 0.5028 0.6943 0.8618 1.1096 1.2979 1.5103 1.9636 2.2656 2.4536
+ 0.2243 0.3890 0.6784 0.9140 1.3747 1.5441 1.8364 2.1029 2.3665 2.5412
+ 0.3574 0.4900 0.6644 0.9247 1.4299 1.6309 1.8339 2.0677 2.3399 2.5100
+ 0.2049 0.3166 0.4561 0.6597 0.8180 1.2661 1.8436 2.0563 2.3772 2.5481
+ 0.5215 0.6958 0.8630 1.0637 1.2808 1.4517 1.8167 2.0873 2.3548 2.5445
+ 0.1920 0.2967 0.4607 1.0024 1.2106 1.4533 1.7725 1.9554 2.4279 2.6166
+ 0.1961 0.3152 0.4661 0.7012 1.2565 1.4225 1.7517 2.0288 2.2747 2.7254
+ 0.2361 0.3248 0.7769 0.9618 1.1800 1.3709 1.7349 2.0322 2.3025 2.5319
+ 0.1507 0.2299 0.3858 0.7118 0.9747 1.3748 1.6886 1.8747 2.0889 2.6390
+ 0.2091 0.2900 0.7508 1.0339 1.4566 1.6563 1.9297 2.1663 2.4403 2.6298
+ 0.2137 0.3071 0.5570 0.7583 0.9379 1.3126 1.6630 1.9210 2.2291 2.4279
+ 0.2668 0.3438 0.6963 1.2648 1.4357 1.6588 1.9702 2.1460 2.5448 2.7099
+ 0.2970 0.4487 0.6175 0.8613 1.0050 1.3940 1.8076 1.9894 2.5570 2.7420
+ 0.4762 0.7388 1.2244 1.4707 1.7598 1.9985 2.2389 2.4028 2.6244 2.7586
+ 0.1499 0.2423 0.3628 0.6927 1.4805 1.6390 1.8853 2.0631 2.5860 2.7761
+ 0.2697 0.3585 0.5474 0.7091 0.9169 1.7121 2.0156 2.2277 2.5745 2.7249
+ 0.2710 0.3558 0.5486 1.1625 1.3530 1.5316 1.8323 2.0087 2.5642 2.7372
+ 0.2633 0.3563 0.5213 0.9584 1.4262 1.5873 1.8383 2.1652 2.3959 2.5696
+ 0.3848 0.4786 0.7087 1.0104 1.1913 1.3713 1.7547 1.9247 2.2100 2.5835
+ 0.2271 0.2927 0.6275 0.8523 1.5725 1.7959 1.9890 2.1862 2.4165 2.5896
+ 0.2181 0.3144 0.4475 0.6284 0.7800 1.3021 2.0094 2.1909 2.5936 2.7693
+ 0.3266 0.4408 0.6486 0.9220 1.0907 1.3106 1.8829 2.0679 2.3198 2.6010
+ 0.1837 0.2617 0.4122 1.3626 1.5493 1.7333 2.0105 2.2742 2.5643 2.7559
+ 0.2976 0.4485 0.7169 0.9778 1.3471 1.5838 1.8802 2.1805 2.5009 2.7376
+ 0.2328 0.3312 0.5098 1.0009 1.1994 1.3786 1.6356 1.9070 2.2851 2.4968
+ 0.2794 0.3826 0.5417 0.7564 0.9045 1.0999 1.6871 2.0486 2.3119 2.5024
+ 0.4184 0.6037 0.7821 0.9673 1.2126 1.3783 1.6892 1.9655 2.2025 2.4035
+ 0.3731 0.4464 0.7658 1.2070 1.3602 1.5675 1.8835 2.0683 2.5124 2.6904
+ 0.2314 0.3515 0.4796 0.6901 0.8348 1.0281 1.9635 2.2122 2.4882 2.7636
+ 0.2659 0.3439 0.5942 0.7509 0.9296 1.5571 1.8076 1.9998 2.5551 2.7010
+ 0.1708 0.2477 0.4112 0.6832 1.1844 1.3761 1.7284 1.9579 2.1848 2.3383
+ 0.5537 0.7504 1.0120 1.2259 1.4638 1.6713 1.9487 2.1717 2.4595 2.6438
+ 0.3007 0.3927 0.6536 0.8457 1.0508 1.5774 1.8298 2.0251 2.2992 2.4802
+ 0.2909 0.3741 0.6808 0.9210 1.0837 1.3417 1.6104 1.8392 2.1065 2.3272
+ 0.1559 0.2363 0.4173 0.6932 1.1096 1.5574 1.8077 2.0036 2.5394 2.7590
+ 0.1930 0.3360 0.5244 0.7608 1.1001 1.2636 1.6611 1.9009 2.1969 2.7229
+ 0.2266 0.2989 0.5046 1.2690 1.4832 1.6462 1.8863 2.0682 2.5529 2.7282
+ 0.1862 0.2944 0.4228 0.6137 1.3243 1.5472 1.7476 2.1261 2.3642 2.5747
+ 0.3062 0.3861 0.7636 1.0288 1.1818 1.3707 1.5442 1.9270 2.3840 2.5222
+ 0.4173 0.6864 0.8879 1.0616 1.2781 1.5609 1.9556 2.2101 2.5238 2.7190
+ 0.3728 0.4653 0.7041 0.8740 1.0346 1.4889 1.7781 1.9515 2.4864 2.6485
+ 0.2813 0.3594 0.8817 1.1702 1.3356 1.6531 1.8890 2.1551 2.5163 2.6695
+ 0.2129 0.2927 0.8019 1.0801 1.3011 1.5868 1.8151 1.9916 2.2430 2.5826
+ 0.2763 0.3397 0.5720 1.1786 1.4411 1.5882 1.8458 2.0294 2.2667 2.5668
+ 0.1668 0.2484 0.3929 1.0800 1.5384 1.6789 1.9590 2.1457 2.5674 2.7524
+ 0.2995 0.3769 0.6437 1.0753 1.2686 1.4490 1.7452 1.9392 2.1687 2.3929
+ 0.4754 0.7185 1.0696 1.3232 1.6208 1.8916 2.2074 2.4001 2.6215 2.7522
+ 0.2396 0.3865 0.5341 0.7449 1.3217 1.5461 1.7311 2.0783 2.3293 2.5500
+ 0.1545 0.2397 0.4226 0.7028 0.9431 1.1939 1.5477 1.7785 2.2104 2.7243
+ 0.1211 0.1950 0.6744 0.9174 1.1811 1.3760 1.7171 1.9820 2.4052 2.7037
+ 0.2379 0.3559 0.5236 0.7186 0.8601 1.1713 1.7691 1.9739 2.5457 2.7405
+ 0.2756 0.3626 0.7125 0.9564 1.1504 1.4734 1.7712 1.9533 2.4390 2.6282
+ 0.1573 0.2128 0.3684 0.5485 1.1972 1.4148 1.6668 2.0137 2.3316 2.6492
+ 0.2152 0.4097 0.6075 0.7645 1.0899 1.2720 1.5197 1.9796 2.2226 2.5092
+ 0.2094 0.3517 1.0980 1.3858 1.6697 1.8943 2.1301 2.3170 2.5542 2.7263
+ 0.3657 0.5074 0.6979 0.8770 1.0621 1.2242 1.6532 1.9868 2.2156 2.4252
+ 0.2063 0.3047 0.4373 0.6036 0.7849 0.9333 1.4718 2.1231 2.4097 2.6422
+ 0.1871 0.2826 0.4687 0.8386 1.1107 1.3254 1.6731 1.8728 2.0899 2.2691
+ 0.4570 0.5569 0.7986 1.0155 1.1670 1.4981 1.8278 2.0003 2.4496 2.6444
+ 0.1654 0.2923 0.4563 0.6790 0.8698 1.1095 1.5515 1.7714 2.5700 2.7883
+ 0.2731 0.4047 0.5913 0.7642 0.9379 1.0969 1.6068 2.2200 2.4888 2.6782
+ 0.2963 0.4236 0.8892 1.1434 1.3432 1.5271 1.7864 2.0219 2.3320 2.5511
+ 0.2774 0.6153 1.0715 1.4189 1.6966 1.9796 2.2484 2.4369 2.6544 2.7978
+ 0.1708 0.2326 0.4438 0.7063 1.1451 1.3412 1.7174 1.9140 2.4104 2.5671
+ 0.3115 0.3929 0.6620 1.1595 1.4066 1.5686 1.9349 2.1353 2.3945 2.6164
+ 0.2994 0.3921 0.5941 0.9092 1.1236 1.2931 1.8266 2.2400 2.4430 2.6005
+ 0.2502 0.3841 0.5481 1.0097 1.2237 1.4210 1.6994 1.8953 2.5762 2.7623
+ 0.3844 0.5541 0.8956 1.0661 1.2330 1.4020 1.7266 2.0392 2.4379 2.6744
+ 0.2518 0.3118 0.6078 1.2466 1.5919 1.7331 1.9383 2.1363 2.3606 2.5907
+ 0.1817 0.2735 0.4265 1.1565 1.3851 1.5703 1.8070 2.0188 2.4711 2.6381
+ 0.2863 0.4920 0.8994 1.2784 1.6748 1.9158 2.2055 2.3899 2.6222 2.7825
+ 0.2538 0.4525 0.6716 0.8342 1.0514 1.2324 1.5285 2.1257 2.3801 2.5957
+ 0.1656 0.2557 0.4754 0.7854 1.0174 1.5229 1.8935 2.2250 2.5797 2.7560
+ 0.2126 0.3152 0.4614 0.8558 1.2696 1.4309 1.9473 2.1460 2.4060 2.6131
+ 0.2416 0.3557 0.5318 0.7569 1.1107 1.2908 1.5501 2.2244 2.6016 2.7591
+ 0.1803 0.2894 0.4074 0.5762 0.7294 0.9863 1.8692 2.1232 2.4965 2.7801
+ 0.1602 0.2526 0.4135 0.6285 0.7946 1.4507 1.7804 2.0623 2.5660 2.7607
+ 0.3068 0.3879 0.5859 0.9980 1.2839 1.4331 1.8261 2.1118 2.3295 2.5154
+ 0.2346 0.3123 0.5079 0.8178 1.3797 1.8754 2.0980 2.2540 2.5021 2.6569
+ 0.1344 0.1887 0.3386 0.5159 0.7964 1.1077 1.6814 1.9596 2.3477 2.7134
+ 0.3438 0.4163 0.6379 1.0504 1.5410 1.7150 1.8916 2.1385 2.3578 2.5197
+ 0.1218 0.1738 0.3110 0.4857 0.9677 1.3695 1.7335 1.9952 2.4210 2.7446
-10 128
- -0.0147 -0.0342 -0.0543 0.1092 0.0408 -0.0072 -0.0648 -0.1229 -0.1947 0.0071
- -0.0091 -0.0209 -0.0622 0.0787 0.0346 0.0103 -0.0146 -0.0648 -0.1321 -0.1730
- -0.0035 -0.0175 -0.0320 0.0123 -0.0854 0.0104 0.0407 -0.0322 0.1128 0.0657
- -0.1010 -0.1327 0.1325 0.0642 0.0262 -0.0060 -0.0448 -0.0250 -0.0061 0.0206
- -0.0444 -0.0613 0.1151 0.0487 0.0388 0.0108 0.0018 -0.0171 -0.0965 -0.0875
- 0.0810 0.0739 0.0370 -0.0279 -0.0963 0.0392 0.0869 0.0196 0.0511 0.0172
- 0.0537 0.0361 0.0082 0.0990 0.0353 -0.0108 -0.0637 0.0517 -0.0145 -0.0649
- -0.0602 -0.0754 0.0489 -0.0163 0.0956 0.0965 0.0648 0.0463 0.0058 -0.0133
- 0.0293 0.0603 0.0167 -0.0224 0.0780 0.0279 -0.0613 -0.1258 0.0449 0.0035
- 0.0174 -0.0102 0.0084 -0.0415 -0.0645 0.0824 -0.0379 -0.1126 0.0116 -0.0417
- 0.0965 0.0945 0.0524 -0.0058 -0.0460 -0.0694 -0.0885 -0.1297 -0.0643 -0.0695
- -0.0419 -0.0705 0.0086 -0.0729 -0.0954 0.0065 -0.0522 -0.0364 -0.0704 -0.0963
- 0.0074 -0.0003 -0.0243 -0.0588 -0.1308 -0.1649 0.0996 0.0375 -0.0314 -0.0634
- 0.0201 0.0046 -0.0083 0.0845 0.1524 0.1054 0.0915 0.0709 0.0431 0.0279
- -0.0307 -0.0226 0.0579 -0.0178 -0.0520 -0.1040 -0.0811 -0.1101 -0.0603 0.0067
- -0.0023 -0.0054 -0.0283 -0.0277 -0.0502 -0.0847 -0.1628 -0.2278 0.1106 0.0762
- 0.0401 0.0611 0.0217 -0.0237 -0.0896 -0.1339 0.0377 -0.0015 -0.0794 0.0654
- -0.0696 -0.0948 0.0192 -0.0333 -0.0427 -0.0568 0.0680 0.0767 0.0298 0.0188
- 0.0145 0.0059 0.1447 0.1026 0.0129 -0.0477 -0.1243 0.0059 0.0641 0.0344
- -0.0260 -0.0537 0.0497 0.0548 0.0185 -0.0202 -0.1208 -0.1595 0.0110 -0.0004
- 0.0073 -0.0061 -0.0162 0.1075 -0.0093 -0.0955 0.0781 0.0390 0.0147 0.0127
- -0.0463 -0.0607 -0.0995 -0.0043 -0.0090 -0.0712 0.0648 -0.0114 0.1143 0.0906
- -0.0273 -0.0416 -0.0987 -0.0978 0.0281 -0.0438 0.0545 0.0092 -0.0516 -0.0590
- -0.0094 -0.0341 0.0031 0.0376 -0.0002 0.0182 0.1845 0.1550 0.1127 0.0704
- 0.0164 0.0066 -0.0223 0.1193 0.0670 0.0326 -0.0688 -0.1218 0.1219 0.0858
- -0.0393 0.1155 0.1050 0.0798 0.0923 0.0493 0.0349 0.0172 -0.0108 -0.0101
- 0.0113 -0.0043 -0.0206 -0.0938 -0.1470 -0.0184 0.2036 0.1436 0.0885 0.0520
- -0.0046 -0.0146 -0.0485 0.0874 0.0728 0.0130 0.0787 0.0179 0.1454 0.1057
- 0.0147 -0.0202 0.0980 0.0341 0.0089 0.1605 0.1168 0.0674 0.0047 -0.0356
- -0.0184 -0.0422 -0.0348 -0.0878 -0.1492 -0.0190 -0.1073 0.0643 0.0785 0.0469
- 0.0017 0.0010 -0.0188 -0.0389 0.1461 0.0829 0.0106 -0.0419 -0.1335 0.0455
- -0.0141 -0.0342 -0.0581 -0.1275 0.0029 0.0788 0.0009 -0.0857 -0.1333 0.0701
- 0.0246 0.0300 -0.0139 -0.1013 0.0553 -0.0046 0.1340 0.0891 -0.0000 -0.0099
- 0.0193 0.0188 -0.0304 -0.0935 -0.0724 -0.1278 -0.0981 0.1049 0.0213 -0.0324
- 0.0122 0.0135 0.0045 0.0129 0.0863 0.0323 0.1000 0.0333 -0.0765 -0.1290
- -0.0254 -0.0335 0.0242 -0.0151 0.0392 -0.0143 -0.0011 0.1281 0.0632 0.0992
- 0.0300 0.0122 0.0710 0.0054 0.0555 -0.0009 -0.1127 0.0667 0.0725 0.0077
- 0.0120 0.0208 -0.0224 -0.0305 -0.0824 -0.1289 0.0934 0.0116 0.1218 0.1015
- 0.0553 0.0466 0.0179 0.0856 0.0036 -0.0286 -0.0862 -0.1354 0.0003 -0.0427
- -0.0769 -0.1097 -0.0330 0.0004 0.0368 0.0187 -0.0178 -0.0549 -0.0975 -0.0481
- -0.0056 -0.0125 -0.0498 -0.1224 0.0879 0.0490 0.0010 -0.0504 -0.1282 -0.1455
- 0.0345 0.0252 0.0639 0.0436 0.0018 -0.0101 -0.0604 -0.1014 -0.1625 -0.1853
- 0.0459 0.0275 -0.0592 0.0171 -0.0051 -0.0655 0.0070 -0.0071 -0.0847 -0.0419
- 0.0117 -0.0019 -0.0107 -0.0198 -0.0686 -0.1346 -0.2551 0.0653 0.0546 0.0169
- 0.0419 0.0643 0.0439 -0.0183 0.1345 0.1117 0.0709 0.0597 0.0206 -0.0108
- 0.0175 0.0235 -0.0410 -0.0876 0.0677 -0.0024 -0.0858 0.0449 -0.0431 -0.0878
- 0.0360 0.0324 -0.0748 0.0262 -0.0073 0.0055 -0.0062 -0.0367 0.0630 0.0306
- -0.0206 -0.0325 -0.0958 0.0271 0.0899 0.0260 0.0175 -0.0376 -0.0852 0.0785
- 0.0040 -0.0218 0.0395 -0.0094 0.0329 -0.0232 0.0155 0.1310 0.0027 -0.0487
- -0.0006 -0.0102 -0.0410 -0.0832 -0.1154 -0.1505 -0.0286 -0.1005 0.0503 0.0202
- 0.0515 0.0546 0.0414 -0.0274 -0.0806 -0.0317 -0.1043 -0.0968 0.1308 0.0757
- 0.0393 0.0437 0.0019 0.1016 0.1144 0.0619 -0.0133 -0.0693 -0.0811 -0.0591
- -0.0250 -0.0349 -0.0530 -0.0880 -0.0059 -0.0552 -0.0947 0.0038 -0.0871 0.0250
- 0.0169 0.0118 -0.0051 -0.0363 -0.0630 -0.1078 -0.2100 -0.0420 -0.0627 -0.1092
- 0.0626 0.0878 0.0130 -0.0802 -0.0096 -0.0056 -0.0393 0.0229 0.0195 0.0029
- -0.0155 -0.0260 -0.0791 -0.0859 0.0490 -0.0029 -0.0876 0.0587 0.1117 0.0731
- -0.0394 -0.0617 -0.0876 0.0632 0.0291 0.1289 0.0436 0.0422 0.0817 0.0397
- -0.0230 -0.0479 -0.0184 0.1379 0.1331 0.1046 0.0278 -0.0011 -0.0559 -0.0687
- -0.0196 0.1098 0.1117 0.0376 -0.0293 -0.0616 -0.0564 -0.0511 0.0229 0.0231
- -0.0025 -0.0207 -0.0026 -0.0006 -0.0382 0.0972 0.0288 -0.0394 -0.1751 -0.0940
- -0.0078 -0.0191 -0.0743 0.1149 0.0452 0.0087 -0.0896 0.1423 0.1322 0.0749
- -0.0181 -0.0360 -0.0840 0.1022 0.0532 -0.0138 -0.0959 0.0211 -0.0489 -0.0490
- 0.0586 0.0749 0.0183 -0.0150 0.0687 0.0499 0.0014 -0.0256 -0.0728 -0.0978
- -0.0099 -0.0323 -0.0548 -0.1367 -0.0033 0.0847 -0.0199 0.0653 0.0165 -0.0188
- 0.0489 0.0751 0.0644 0.0106 -0.0386 -0.0836 -0.0121 0.1574 0.0978 0.0558
- 0.0018 -0.0064 -0.0594 -0.1062 -0.1658 0.0249 0.0265 -0.0343 -0.0636 -0.0941
- 0.0094 0.0094 -0.0222 -0.0419 -0.0699 -0.1009 0.0095 -0.0619 -0.1466 -0.1801
- -0.0038 -0.0213 -0.0099 -0.0550 -0.1430 -0.0352 -0.0108 -0.0902 -0.0990 0.0438
- -0.0391 -0.0655 0.0185 -0.0434 -0.0090 0.0218 -0.0391 -0.0985 0.1176 0.0854
- -0.0037 -0.0168 -0.0630 -0.1078 -0.1575 0.0795 0.0005 -0.0563 0.1105 0.0628
- 0.0010 -0.0107 -0.0399 -0.0046 -0.0590 -0.0576 0.1455 0.0691 -0.0479 -0.0921
- -0.0334 0.0136 0.0594 -0.0662 0.0088 -0.0102 0.0221 0.0111 -0.0275 -0.0150
- 0.0677 0.0726 0.0744 0.0114 -0.0403 -0.0757 0.0324 0.0259 -0.0566 -0.0819
- 0.0346 0.0153 0.1297 0.0306 -0.0479 0.0364 -0.0078 -0.0478 0.0124 -0.0244
- 0.0115 0.0970 0.0403 0.0109 0.0545 0.0404 0.0300 0.0262 0.1282 0.0955
- 0.0053 0.0051 -0.0227 -0.0214 0.0723 -0.0227 0.0111 -0.0319 0.0260 -0.0074
- 0.0060 0.0108 -0.0224 0.0366 -0.0504 -0.1083 -0.0260 -0.1100 0.0095 -0.0294
- -0.0075 -0.0363 0.0145 -0.0409 -0.0557 0.1570 0.0774 0.0589 0.1174 0.0630
- -0.0039 0.1001 0.0552 -0.0343 0.0165 -0.0304 -0.0281 -0.0452 -0.0904 -0.0860
- 0.0687 0.0558 0.0365 0.0904 0.0077 -0.0396 0.0275 -0.0518 0.0720 0.0693
- -0.0056 -0.0106 -0.0653 -0.0398 0.1528 0.0959 0.0259 0.0028 -0.0201 -0.0592
- 0.0228 0.0126 -0.0195 -0.0931 -0.0081 0.1680 0.0867 0.0238 -0.0413 -0.0824
- 0.0140 0.0052 -0.0176 -0.0877 0.1051 0.1180 0.0579 0.0076 0.0918 0.0552
- 0.0079 -0.0273 0.0938 0.0138 -0.0477 -0.0275 -0.1236 0.0544 -0.0374 -0.0774
- 0.1175 0.1285 0.0868 0.0706 0.0503 0.0360 0.0051 0.0021 0.0023 -0.0142
- -0.0259 -0.0587 0.0105 0.0977 0.0397 -0.0015 0.0459 -0.0265 -0.0094 0.0595
- 0.0255 0.0087 -0.0292 0.1515 0.0767 0.0595 0.0743 0.0298 -0.0019 -0.0197
- -0.0447 -0.0722 -0.0853 -0.0591 -0.0891 0.0931 0.0636 0.0226 -0.0157 -0.0251
- -0.0092 -0.0182 0.0648 0.0048 -0.0966 -0.1542 -0.0521 0.0755 0.0455 0.0445
- 0.0052 0.0013 0.0210 0.0050 -0.0444 0.1106 0.0641 -0.0159 -0.0881 0.0951
- -0.0251 -0.0396 -0.0752 -0.1097 0.1184 0.0701 0.1276 0.1034 0.0433 0.0328
- -0.0100 -0.0260 -0.0665 -0.0324 -0.0670 -0.1111 0.0604 0.1558 0.0849 0.0487
- 0.0406 0.0299 0.0547 -0.0097 -0.0525 0.1006 -0.0051 0.1272 0.0622 0.0121
- -0.0324 -0.0434 0.1005 0.0069 -0.0719 -0.0116 0.0058 0.0594 0.1237 0.0979
- 0.0683 0.0655 0.0314 0.0748 0.0075 -0.0157 0.1503 0.0950 0.0241 0.0222
- -0.0270 -0.0390 -0.0800 -0.0240 -0.0651 -0.1203 0.0580 -0.0217 -0.1168 0.0590
- 0.0328 0.0449 0.0089 0.0301 -0.0706 0.0363 0.0645 -0.0154 -0.0448 -0.1029
- -0.1610 0.0975 0.0215 -0.0454 -0.0162 -0.0153 -0.0302 -0.0079 -0.0343 -0.0058
- -0.0236 -0.0502 0.0672 0.1472 0.0698 0.0498 -0.0108 0.0730 0.0404 0.0148
- 0.0177 0.0265 -0.0316 -0.0609 -0.0647 -0.0342 -0.1142 0.2303 0.1815 0.1102
- -0.0200 -0.0553 0.0420 0.0520 -0.0653 0.0356 0.0789 0.0216 -0.0054 -0.0371
- 0.0088 0.0187 -0.0480 -0.1326 0.0108 -0.0327 -0.0693 -0.1156 0.0043 -0.0012
- 0.0082 -0.0001 -0.0623 0.0107 -0.0475 0.1223 0.1220 0.0633 0.0459 0.0061
- -0.0063 -0.0261 0.0375 -0.0171 0.0934 0.0337 -0.0625 -0.0185 -0.0642 -0.1047
- -0.0432 -0.0761 -0.0888 -0.0203 0.0424 -0.0029 -0.0429 0.0580 0.0191 -0.0331
- -0.0064 -0.0098 -0.0292 -0.0442 -0.0311 -0.0692 -0.1032 -0.1564 -0.2260 0.0398
- -0.0265 -0.0424 -0.1001 0.0014 -0.0637 -0.0063 -0.0525 -0.1065 0.0004 -0.0410
- 0.0145 -0.0077 0.1417 0.0800 0.0254 -0.0140 0.1199 0.0921 0.0366 0.0291
- 0.0276 0.0304 0.0420 0.0427 0.0185 -0.0148 -0.0656 -0.1486 -0.1264 0.0927
- -0.0292 -0.0425 -0.0832 -0.0815 0.1040 0.0397 -0.0615 -0.1320 0.0701 0.0516
- -0.0204 -0.0391 -0.0536 0.0029 -0.0463 -0.1175 -0.0620 0.0196 -0.0771 -0.1071
- 0.0006 0.0056 -0.0265 0.0111 0.0419 -0.0209 0.1261 0.0562 -0.0380 0.1026
- -0.0772 0.1495 0.1096 0.0310 -0.0119 0.0310 0.1149 0.1370 0.0874 0.0617
- 0.0297 0.0336 0.0141 0.0530 0.0093 -0.0372 0.0304 -0.0387 -0.1447 0.0011
- -0.0179 -0.0367 -0.0817 0.1134 0.0707 0.0120 -0.0452 -0.1060 0.0488 -0.0015
- 0.0409 0.0356 0.0243 0.0752 0.0088 0.1064 0.0295 -0.0369 0.0841 0.0314
- 0.0338 0.0050 0.1358 0.1160 0.0815 0.0488 0.0167 -0.0140 -0.0441 -0.0683
- 0.0163 0.0106 -0.0497 0.0089 0.0889 0.0220 0.0444 0.1229 0.0578 0.0128
- 0.0217 0.0172 -0.0237 -0.0151 -0.1139 0.0031 -0.1051 -0.0274 0.0003 -0.0414
- -0.0138 -0.0220 -0.0676 -0.1531 -0.0121 -0.0520 0.0733 0.0472 0.0732 0.0526
- -0.0100 -0.0324 0.1059 0.0599 0.1064 0.0761 0.0101 -0.0393 0.0684 0.0503
- -0.0263 -0.0505 -0.0910 0.0919 0.0648 0.0128 0.1304 0.0943 0.0113 -0.0037
- 0.0138 -0.0012 0.0282 -0.0654 -0.1621 0.0828 0.0696 0.0341 -0.0190 -0.0422
- -0.0095 -0.0187 -0.0493 -0.0305 0.0267 -0.0225 -0.1006 -0.1347 -0.1064 -0.1385
- -0.0001 0.0029 -0.0226 -0.0455 -0.0045 -0.0561 -0.0208 -0.0381 -0.0653 0.1681
- 0.0163 0.0612 0.0337 -0.0040 0.0303 -0.0352 -0.0824 0.0184 -0.0772 0.0627
- -0.1619 0.0289 0.0324 0.0509 0.0320 0.0320 0.0065 0.0320 0.0499 0.0672
- -0.0476 -0.0707 -0.0346 0.0537 -0.0334 -0.0995 -0.1149 -0.0176 0.0757 0.0611
-
+10 256
+ -0.0388 -0.0448 0.0718 0.0236 0.0461 -0.0074 -0.0045 -0.0744 -0.0979 0.0917
+ -0.0594 -0.0950 0.0403 -0.0422 -0.1153 -0.0259 0.0572 0.0667 0.0210 0.0034
+ 0.0260 0.0664 0.0204 -0.0106 0.0517 -0.0060 0.0094 -0.0223 -0.0314 0.1428
+ 0.0602 0.0440 0.1044 0.0179 -0.0112 0.0233 -0.0048 -0.0344 -0.0391 -0.0767
+ -0.0116 -0.0184 -0.0305 0.0507 0.0122 -0.0346 -0.0882 -0.1275 -0.1656 -0.1946
+ -0.0526 -0.0878 -0.0824 0.0520 -0.0120 -0.0057 0.1112 0.0669 0.0067 -0.0098
+ 0.0071 -0.0051 0.0062 0.0560 0.0114 -0.0544 -0.0607 -0.0451 -0.1365 -0.0438
+ -0.0277 -0.0398 -0.0767 0.0732 0.0210 -0.0061 -0.0369 -0.0650 0.1656 0.1101
+ 0.0056 0.0015 -0.0253 -0.0701 0.0668 -0.0055 -0.0622 0.1525 0.1171 0.0863
+ 0.0150 0.0106 0.1502 0.0880 0.0200 0.0028 0.0883 0.0589 0.0784 0.0597
+ 0.0230 0.0752 0.0179 -0.0042 -0.0458 -0.0944 -0.0303 -0.0876 0.0190 -0.0269
+ 0.0153 0.0049 -0.0402 0.1146 0.0930 0.0521 0.0743 0.0323 0.1185 0.0858
+ -0.0358 -0.0526 -0.0997 0.0474 -0.0109 0.1087 0.0361 0.0946 0.1016 0.0670
+ -0.0397 -0.0659 0.0503 0.1424 0.0577 0.0229 0.0062 0.0025 -0.0406 -0.0565
+ 0.0821 0.0758 0.0284 0.1126 0.0833 0.0719 0.0383 0.0052 0.0157 -0.0119
+ 0.0256 0.0125 -0.0695 0.0534 0.1412 0.0884 0.0229 0.0130 -0.0444 -0.0414
+ -0.0088 -0.0328 -0.0320 -0.1081 0.0927 0.1051 -0.0099 0.0969 0.0100 -0.0058
+ -0.0025 -0.0244 0.0835 0.0321 -0.0325 0.0494 0.1347 0.0625 -0.0293 -0.0592
+ 0.0038 -0.0009 0.0106 0.0156 0.0715 0.0170 -0.0870 -0.1503 0.1522 0.1104
+ -0.0253 -0.0361 -0.1018 -0.0250 0.0433 -0.0489 0.1182 0.0743 -0.0216 -0.0276
+ -0.0168 -0.0358 0.0289 -0.0741 -0.0962 -0.0068 -0.0544 -0.0652 -0.0321 -0.0721
+ -0.0226 -0.0391 -0.0844 0.0472 0.0024 0.0820 -0.0133 -0.0865 0.0510 0.0116
+ -0.0092 -0.0282 -0.1107 0.0255 -0.0160 -0.0710 -0.0995 0.0280 -0.0506 -0.0943
+ -0.0169 -0.0451 0.0494 0.0563 -0.0726 -0.0805 0.0071 -0.0479 0.0125 0.0353
+ 0.0400 0.0335 0.0513 -0.0390 0.0993 0.1362 0.0582 0.0228 0.0801 0.0381
+ -0.0435 -0.0852 -0.0297 -0.0220 0.0017 -0.0300 -0.0469 0.0107 -0.0733 -0.1121
+ 0.0107 -0.0078 0.0390 -0.0022 -0.0670 0.0143 0.1181 0.0658 0.0905 0.0827
+ 0.0262 0.0167 -0.0136 -0.0402 -0.1175 0.0265 -0.0788 -0.1620 0.0617 0.0100
+ -0.0004 -0.0033 0.0011 0.0024 0.0542 -0.0120 0.0617 0.0104 -0.1479 -0.0268
+ -0.0113 -0.0310 -0.0338 -0.0785 -0.0506 0.0085 -0.1328 -0.0275 0.0604 0.0164
+ 0.0089 -0.0152 0.0384 -0.0395 0.0537 0.1398 0.0844 0.0274 -0.0464 -0.0756
+ 0.0313 0.0513 -0.0211 0.0925 0.0614 -0.0120 0.0034 0.0615 -0.0177 -0.0265
+ 0.0926 0.1094 0.0393 -0.0095 -0.0289 -0.0628 -0.1154 -0.1121 -0.0476 -0.0326
+ 0.0140 0.0076 -0.0226 -0.1038 0.0930 0.0146 0.1595 0.1176 0.0324 0.0084
+ 0.0159 -0.0051 0.0695 0.0338 -0.0356 -0.0735 -0.0174 -0.0850 0.1175 0.1095
+ -0.0267 0.0451 -0.0052 -0.0474 0.0154 -0.0354 -0.0629 0.0260 -0.0367 -0.0515
+ -0.0509 -0.0693 0.0258 -0.0142 0.1054 0.0702 0.0208 -0.0118 -0.0680 -0.0829
+ 0.0409 0.0877 0.0209 -0.0767 0.0233 -0.0175 -0.0045 0.0527 0.0414 0.0178
+ -0.0046 -0.0006 -0.0411 -0.0808 0.0017 -0.0707 0.0037 -0.0459 -0.1297 -0.1546
+ 0.0414 0.0461 0.0061 -0.0232 -0.0280 -0.0416 -0.0750 -0.1091 -0.1349 -0.1495
+ 0.0541 0.0601 0.0057 0.0587 0.0144 -0.0310 -0.0869 -0.1062 0.0846 0.0406
+ 0.0069 -0.0000 -0.0257 -0.0575 0.1055 0.0461 -0.0323 -0.0816 -0.1463 -0.1572
+ -0.0097 -0.0354 0.0836 0.1357 0.0654 0.0264 -0.0403 -0.0792 0.0919 0.0569
+ 0.0008 -0.0217 -0.0332 -0.1156 -0.0393 0.1228 0.0227 -0.0602 0.0424 -0.0152
+ -0.0364 -0.0624 -0.0497 -0.0335 0.0117 -0.0460 0.0067 0.1152 0.0671 0.0290
+ -0.0337 -0.0518 -0.0813 -0.1195 0.0599 0.0157 -0.0503 -0.0989 0.1165 0.0687
+ 0.0253 0.0057 -0.0764 0.0023 0.0192 -0.0261 0.0601 0.1356 0.0598 0.0323
+ 0.0189 0.0218 -0.0202 -0.0381 -0.1006 -0.1560 0.0613 -0.0097 0.0979 0.0609
+ -0.0142 -0.0133 -0.0759 -0.1357 0.1003 0.0358 0.0450 0.0044 -0.0260 -0.0464
+ 0.0124 -0.0185 0.0735 -0.0084 -0.0601 0.1034 0.0291 -0.0149 -0.0901 -0.1366
+ 0.0449 0.0455 0.0167 0.0931 0.0522 0.0670 0.0097 0.0929 0.0973 0.0438
+ 0.0148 -0.0071 -0.0007 -0.0872 -0.1297 -0.0398 0.0365 -0.0338 0.0935 0.0610
+ 0.0120 -0.0069 -0.0250 0.1371 0.0726 0.0473 0.1336 0.0824 0.0012 -0.0160
+ -0.0027 -0.0169 -0.0917 0.0203 0.0322 -0.0168 -0.0553 -0.1007 -0.1454 -0.0398
+ -0.0229 -0.0345 -0.0613 -0.1092 -0.0169 -0.1084 0.0013 -0.0496 0.0759 0.0375
+ -0.1027 0.0813 0.0274 -0.0446 0.0291 0.0730 0.0888 0.0920 0.0418 0.0258
+ 0.0094 0.0150 -0.0029 -0.0657 0.0064 -0.0338 -0.0987 -0.1261 -0.2111 0.0080
+ -0.0527 -0.0809 -0.0615 0.0629 0.0247 -0.0117 -0.0375 -0.0983 -0.1161 0.0363
+ -0.0122 -0.0241 -0.0358 0.0366 -0.0649 0.0340 -0.0435 -0.1040 -0.0322 -0.0986
+ 0.0323 0.0211 -0.0295 0.0395 -0.0105 0.0962 0.0085 0.1202 -0.0126 -0.0469
+ -0.0020 -0.0253 0.0264 0.0252 0.0297 0.1726 0.0717 0.0233 0.0630 0.0083
+ 0.0135 -0.0013 0.0085 0.0457 0.0088 -0.0351 0.1764 0.1146 0.0020 -0.0597
+ -0.0227 -0.0334 -0.0421 -0.0044 -0.0757 0.0795 -0.0231 -0.0984 -0.1621 0.0463
+ -0.0273 -0.0434 -0.0673 -0.0126 -0.0588 -0.1018 0.0909 0.0158 -0.0870 0.0647
+ -0.0166 -0.0312 -0.0556 0.0506 0.0062 0.1075 0.0855 0.0093 -0.0790 0.0391
+ -0.0117 -0.0187 -0.0503 0.0134 0.1307 0.0798 0.0929 0.0698 0.0204 -0.0055
+ -0.0200 -0.0356 -0.0653 0.1188 0.0827 0.0194 0.0039 0.1140 0.0223 -0.0016
+ 0.0223 0.0111 0.1235 0.0610 -0.0344 -0.0815 -0.1164 0.0621 0.0509 0.0150
+ 0.0199 0.0049 0.0952 0.0228 -0.0305 -0.0948 0.0485 0.0047 -0.0736 0.0405
+ 0.0231 0.0233 -0.0248 -0.0321 -0.0993 0.0885 -0.0713 0.1863 0.1235 0.0713
+ 0.0249 0.0475 0.0049 -0.0892 -0.0521 -0.0863 0.0190 -0.0264 -0.0733 0.0135
+ -0.0037 -0.0098 -0.0392 0.1061 -0.0222 -0.1195 0.1081 0.0681 0.0328 0.0297
+ -0.0241 -0.0356 -0.0728 -0.0665 -0.0698 0.0948 0.1163 0.0435 -0.0436 -0.1014
+ 0.0771 0.0539 -0.0337 0.0032 -0.0319 -0.0779 0.0425 0.0164 -0.0384 -0.0049
+ -0.0200 -0.0463 0.0546 -0.0275 -0.0429 0.0472 -0.0687 -0.1068 0.0804 0.0594
+ 0.0135 0.0061 -0.0170 0.0302 -0.0367 -0.0768 -0.1833 -0.0706 -0.0200 -0.0859
+ -0.0048 -0.0126 -0.0395 -0.0606 0.1435 0.0753 -0.0157 -0.0767 0.0165 -0.0380
+ -0.0149 -0.0204 -0.0725 -0.0124 -0.0883 -0.0702 0.1520 0.0717 -0.0204 -0.0738
+ 0.0340 0.0745 -0.0069 -0.0318 0.1068 0.0534 -0.0164 -0.0689 0.0426 0.0179
+ 0.0014 -0.0081 -0.0400 0.0501 0.0004 0.1115 0.0528 -0.0222 -0.1313 -0.1611
+ -0.0106 -0.0425 0.0444 -0.0028 -0.0166 0.0111 -0.0575 0.1284 -0.0212 -0.0725
+ 0.0282 0.0077 0.1315 0.0882 0.0312 -0.0077 -0.0712 -0.1013 -0.0737 -0.0850
+ -0.0292 0.0692 0.0456 -0.0192 -0.0829 0.0150 0.0265 -0.0351 0.0539 0.0292
+ -0.0331 0.0631 0.0850 -0.0581 0.0346 -0.0001 0.0148 -0.0089 -0.0501 -0.0425
+ 0.0116 0.0046 0.0079 -0.0015 -0.0468 0.0998 0.0300 -0.0461 0.1537 0.0999
+ 0.0032 0.0002 -0.0452 -0.0338 -0.1053 -0.1275 -0.0483 -0.1059 -0.0106 -0.0536
+ -0.0113 -0.0314 0.0692 0.0281 0.1084 0.0542 -0.0305 0.1004 0.0678 0.0481
+ 0.0003 -0.0221 0.0068 -0.0124 -0.0277 -0.0336 0.0188 -0.0611 -0.0034 -0.0429
+ -0.0117 -0.0206 -0.0418 -0.1120 -0.0711 0.1151 0.0821 0.0075 -0.0804 0.0725
+ 0.0551 0.0410 0.0070 -0.0743 -0.1144 -0.0423 -0.0963 -0.0215 0.0208 -0.0200
+ 0.0041 -0.0228 0.0927 -0.0457 0.0246 -0.0148 0.0481 0.0404 0.0448 0.0043
+ -0.0847 0.0707 -0.0843 0.0221 0.0029 -0.0061 -0.0184 0.0230 0.0447 0.0426
+ -0.0222 -0.0323 0.1595 0.0395 -0.0287 -0.0080 -0.0173 0.0058 -0.0209 -0.0226
+ 0.0540 0.0607 -0.0095 0.0112 -0.0452 -0.0210 0.0415 -0.0105 0.1407 0.0959
+ -0.0037 0.1116 0.0705 0.0533 0.0024 -0.0479 0.0717 0.0527 -0.0005 -0.0011
+ -0.0026 0.0051 -0.0290 -0.0568 0.0009 -0.0509 -0.1070 -0.1665 -0.0076 -0.0655
+ 0.0339 0.0105 0.0750 0.0343 -0.0340 -0.0411 -0.0936 -0.1315 0.0368 -0.0143
+ -0.0215 -0.0329 -0.0586 0.0161 0.0272 -0.0148 0.1789 0.1499 0.1179 0.0817
+ 0.0503 0.0347 0.0096 0.1076 0.0179 -0.0307 0.0345 -0.0409 0.0743 0.0439
+ -0.0036 0.0493 -0.0031 -0.0170 -0.0175 -0.0739 0.1356 0.0608 -0.0418 0.0951
+ 0.0215 0.0310 -0.0228 0.0361 -0.0426 -0.0909 0.0439 -0.0347 -0.0571 -0.1173
+ -0.0235 -0.0310 -0.1199 -0.0549 0.0597 0.0012 -0.0510 0.0578 0.0171 0.0116
+ -0.0598 -0.0886 0.0787 0.0318 0.0108 0.1027 0.1086 0.1125 0.0542 0.0265
+ -0.0426 -0.0646 0.1480 0.0959 0.1249 0.0976 0.0418 0.0156 0.0144 -0.0016
+ -0.0235 -0.0382 -0.1079 0.0280 -0.0206 0.1142 0.0320 0.0239 -0.0318 -0.0675
+ -0.0016 0.0222 0.0272 0.0252 0.0338 -0.0477 0.0665 0.0154 0.1355 0.1086
+ -0.0101 -0.0137 -0.0420 -0.0482 0.1416 0.0870 0.0505 -0.0006 -0.0512 0.0936
+ -0.1569 0.1055 0.1190 0.0746 0.0411 0.0168 -0.0210 0.0419 0.0221 0.0431
+ 0.0152 -0.0125 0.0476 -0.0397 0.0069 0.0611 0.0635 0.1735 0.0993 0.0534
+ -0.0022 -0.0066 -0.0532 -0.0873 0.0810 0.1075 0.0196 0.0235 0.1139 0.0593
+ 0.0062 -0.0144 -0.0467 0.1598 0.0495 -0.0300 0.0194 -0.0404 -0.0745 0.0586
+ -0.0168 -0.0323 -0.0615 0.1541 0.1056 0.0652 -0.0137 -0.0426 -0.0922 -0.0914
+ -0.0149 -0.0433 0.0184 0.1162 0.0290 -0.0033 0.0842 0.0476 0.0956 0.0615
+ -0.0114 -0.0179 -0.0480 0.0386 -0.1162 -0.0050 0.0600 0.0172 0.0143 -0.0133
+ 0.0689 0.0513 0.0287 -0.0555 -0.1046 0.1122 0.0781 0.0176 0.0647 0.0174
+ 0.0625 0.0659 0.0702 0.0257 -0.0012 -0.0205 -0.0620 0.1231 0.0775 0.0218
+ -0.0102 -0.0122 -0.0332 -0.0472 0.1187 -0.0061 -0.1233 0.0226 -0.0659 0.0025
+ -0.0102 0.0202 0.0626 -0.0237 -0.0614 -0.1148 -0.1134 -0.0493 -0.0503 -0.0075
+ 0.1071 0.1029 0.0633 0.0141 -0.0262 0.0431 -0.0019 -0.0340 0.0404 0.0031
+ 0.0144 0.0106 -0.0335 -0.0878 -0.1583 0.0113 0.0872 0.0127 -0.0084 -0.0591
+ 0.0794 0.0916 0.0718 0.0780 0.0376 -0.0109 -0.0019 -0.0360 -0.0566 0.0031
+ 0.0191 0.0077 -0.0712 0.0288 -0.0560 0.1080 0.0954 0.0371 0.0745 0.0327
+ 0.0096 0.0078 -0.0305 -0.0175 -0.0446 -0.1118 -0.0038 0.0978 0.0156 0.0947
+ -0.0101 -0.0029 -0.0670 -0.0633 0.0303 -0.0377 0.0431 -0.0198 -0.1065 0.0158
+ -0.0282 -0.0408 -0.0950 -0.0833 -0.0454 0.0033 0.0360 -0.0200 0.0148 0.0266
+ 0.0526 0.0553 0.0023 -0.0188 -0.0704 -0.1153 -0.1703 0.0872 0.0516 0.0149
+ -0.0179 -0.0308 -0.0730 -0.0672 -0.1464 0.0573 -0.0143 -0.0498 0.1191 0.0667
+ -0.0174 -0.0350 0.0788 0.0417 -0.0441 0.0726 -0.0406 0.0883 0.0793 0.0338
+ -0.0303 -0.0683 0.0231 0.0590 -0.0199 0.0672 0.0156 -0.0374 0.0453 0.0159
+ 0.0340 0.0578 0.0174 0.0159 -0.0362 -0.0757 0.0166 -0.0572 -0.1718 -0.0161
+ 0.0039 0.0450 -0.0593 -0.1301 0.0109 -0.0272 -0.0409 -0.0779 -0.0236 -0.0307
+ 0.0466 0.0312 -0.0724 0.0031 -0.0296 0.0298 -0.0261 -0.0711 0.0282 -0.0124
+ -0.0176 -0.0392 -0.0294 -0.0749 -0.0941 0.1079 0.0995 0.0832 0.1013 0.0605
+ -0.0096 0.0756 0.0477 0.0350 0.0018 -0.0151 -0.0235 -0.0593 -0.0920 -0.1054
+ -0.0393 -0.0577 0.1021 0.0403 -0.0168 -0.0715 0.0005 0.1550 0.0879 0.0603
+ 0.0559 0.0707 0.0889 0.0157 0.0541 0.0434 0.1098 0.0817 0.0212 -0.0063
+ 0.0055 -0.0200 0.1100 0.0686 0.0801 0.0559 0.0204 -0.0334 -0.1182 -0.1166
+ 0.0081 -0.0222 0.1073 0.0042 -0.0871 0.0948 0.0225 -0.0241 0.0633 0.0262
+ 0.0119 0.0074 -0.0197 0.1133 0.1274 0.0614 -0.0129 -0.0764 0.0108 0.0575
+ 0.0227 0.0065 0.0389 -0.0127 -0.0162 0.1103 -0.0367 -0.1206 -0.0173 -0.0508
+ 0.0027 0.0009 -0.0322 -0.1115 0.0059 0.0694 0.0132 -0.0082 -0.0915 -0.1071
+ 0.0123 0.0154 -0.0115 0.0046 -0.0207 -0.0214 0.0032 0.0018 0.0887 0.0007
+ -0.0261 -0.0372 -0.0727 -0.0055 -0.0520 -0.0934 0.0742 -0.0020 0.1235 0.0751
+ -0.0202 -0.0451 -0.0673 -0.0873 0.0358 -0.0196 -0.0990 -0.0645 -0.0273 -0.0883
+ 0.0026 -0.0156 0.0572 0.0063 0.0556 -0.0120 -0.0814 -0.0208 -0.0749 -0.1270
+ -0.0152 -0.0171 -0.0509 -0.0738 -0.0393 -0.1016 -0.1152 0.0150 -0.0705 0.0181
+ 0.0355 0.0387 0.0054 -0.0251 0.1005 0.0311 -0.0222 0.1239 0.0214 -0.0383
+ -0.0668 -0.0961 -0.0138 -0.0696 -0.0309 0.0897 0.0006 -0.0121 0.0050 -0.0158
+ -0.0213 -0.0304 -0.0796 0.0420 0.0442 -0.0233 0.0838 0.0063 -0.0538 0.1384
+ 0.0607 0.0670 0.0249 -0.0156 -0.0827 -0.0111 0.1093 0.0296 -0.0748 -0.0928
+ 0.0195 0.0081 0.0014 -0.0833 -0.0118 -0.0699 0.1325 0.0669 -0.0539 -0.0902
+ 0.0422 0.0510 0.0412 0.0121 -0.0389 0.0644 0.0444 -0.0285 -0.0844 0.0853
+ -0.0595 -0.0811 -0.0887 0.0944 0.0162 -0.0248 -0.1339 0.0071 0.0136 0.0028
+ 0.0164 0.0056 -0.0097 -0.0199 -0.0577 -0.1237 -0.0347 0.1165 0.0033 -0.0524
+ 0.0220 0.0049 0.0568 0.0955 -0.0115 0.0285 0.0507 -0.0346 -0.0514 0.0024
+ -0.0178 -0.0391 -0.0172 -0.0296 -0.1495 0.1534 0.0800 0.0355 -0.0072 -0.0191
+ 0.0130 0.0079 -0.0775 -0.0005 0.0734 -0.0248 0.0141 -0.0524 0.0626 0.0581
+ -0.0440 -0.0666 0.0835 0.0739 -0.0005 -0.0499 -0.1459 -0.0911 -0.0096 -0.0069
+ 0.0323 0.0089 0.0803 0.1024 -0.0005 -0.0622 0.0699 0.0420 -0.0407 -0.0735
+ -0.0110 -0.0360 -0.0345 -0.0838 -0.1359 0.0525 -0.0453 0.1013 0.0137 -0.0339
+ -0.0434 -0.0612 -0.0312 -0.1040 0.0681 0.0145 0.0865 0.0704 0.0045 0.0121
+ 0.0034 -0.0158 -0.0055 -0.0778 -0.1483 0.0164 -0.0170 -0.0739 -0.0866 0.0662
+ -0.0968 0.2119 0.0816 -0.0024 0.0375 0.0255 -0.0033 -0.0296 -0.0656 -0.0479
+ -0.0457 -0.0649 -0.0483 0.0569 0.1160 0.0558 0.0225 -0.0176 0.0609 0.0399
+ -0.0173 0.1029 0.0536 0.0695 0.0692 0.0724 0.0393 0.0234 0.0870 0.0553
+ -0.0299 0.0699 0.0550 0.0460 0.1220 0.0777 0.0360 0.0197 -0.0423 -0.0479
+ -0.0417 -0.0790 0.0527 0.0471 0.0153 -0.0500 0.0804 0.0311 -0.0317 -0.0184
+ -0.0432 0.0101 0.0224 -0.0347 0.0555 0.0383 -0.0005 -0.0272 0.0481 0.0076
+ 0.0070 0.0106 -0.0328 0.0059 0.0522 0.0098 0.0969 0.0328 -0.0726 -0.1430
+ -0.0053 -0.0183 -0.0472 -0.1430 -0.0076 -0.0367 -0.0338 0.1050 0.0317 -0.0108
+ 0.0273 0.0090 0.0071 -0.1105 0.0261 -0.0215 -0.0876 -0.0062 -0.0341 -0.0785
+ 0.0284 0.0306 0.0309 -0.0246 -0.0621 0.0037 -0.1238 0.0162 0.1779 0.1051
+ 0.0033 -0.0098 -0.0480 -0.0861 -0.1362 -0.0879 -0.0672 0.1604 0.1077 0.0602
+ 0.0501 0.0267 0.1134 0.0498 0.0442 0.0343 -0.0167 -0.0544 0.0912 0.0536
+ -0.0818 0.0402 0.0457 -0.0075 -0.0406 -0.0795 -0.0359 0.0376 0.0893 0.0937
+ 0.0103 0.0581 0.0677 0.0005 -0.0044 -0.0338 -0.0783 0.0279 -0.0638 0.0775
+ 0.0328 0.0433 -0.0021 0.0455 -0.0944 0.0322 -0.0639 -0.0080 0.0120 -0.0189
+ -0.0304 -0.0500 -0.1071 0.0147 -0.0467 -0.0768 -0.0413 -0.1117 0.0428 -0.0063
+ -0.0004 -0.0284 0.0879 0.0130 -0.0380 0.0813 -0.0077 -0.0430 -0.1511 0.0323
+ -0.0056 -0.0046 -0.0451 -0.0494 -0.0533 -0.1026 -0.0699 -0.1331 0.1455 0.1144
+ -0.0778 0.0095 0.0448 -0.0100 -0.0204 -0.0563 0.0508 0.0375 -0.0393 -0.0554
+ 0.0258 0.0118 0.0304 -0.0427 0.1746 0.0898 0.0301 0.0213 -0.0334 -0.0536
+ -0.0064 -0.0145 -0.0339 -0.0359 -0.1144 -0.0061 0.0106 -0.0553 -0.1766 -0.1111
+ -0.0308 -0.0576 -0.0278 0.0138 0.0289 -0.0456 -0.0787 0.0463 -0.0559 0.0913
+ -0.0134 -0.0302 0.0037 0.0084 0.0422 0.0565 -0.0777 -0.0011 -0.0378 -0.0698
+ 0.0312 0.0088 0.0107 0.1286 0.0325 -0.0259 -0.1204 0.0533 0.0163 -0.0281
+ 0.0081 0.0051 -0.0210 0.0739 -0.0003 -0.0443 -0.0534 0.1833 0.0996 0.0564
+ -0.0218 -0.0323 -0.0635 0.0103 0.0698 -0.0012 -0.1264 -0.1836 0.0253 0.0021
+ 0.0081 -0.0096 0.0159 -0.0331 -0.1037 -0.0597 -0.1698 0.0146 -0.0830 -0.1016
+ 0.0139 -0.0083 0.0608 0.1011 0.1067 0.0648 0.0440 0.0027 -0.0506 0.0620
+ -0.0036 -0.0336 0.0543 -0.0045 0.1037 0.0260 -0.0465 -0.1092 0.0054 -0.0182
+ -0.0209 -0.0052 -0.0076 -0.0015 0.0051 -0.0002 -0.0060 0.0131 -0.0153 -0.0042
+ 0.0269 0.0275 -0.0309 0.0071 -0.0237 -0.0817 -0.1283 -0.0122 -0.0905 0.0830
+ 0.0138 0.0004 -0.0480 0.1240 0.0235 -0.0221 -0.0859 -0.1375 0.0762 0.0252
+ 0.0534 0.0365 0.0059 0.0859 0.0564 0.0279 -0.0076 -0.0388 -0.1173 -0.1383
+ -0.0675 -0.1012 0.0503 -0.0278 -0.0085 -0.0357 -0.0796 0.0349 -0.0241 -0.0060
+ -0.0261 -0.0425 0.0322 -0.0339 -0.1067 -0.1168 -0.1193 0.0233 0.0805 0.0507
+ 0.0481 0.0314 -0.0209 0.0822 0.0370 -0.0209 0.0866 0.0169 -0.0703 0.0637
+ 0.0446 0.0933 0.0269 -0.0322 0.0785 0.0290 0.0313 -0.0117 -0.0655 -0.0783
+ 0.0190 -0.0000 0.0161 -0.0444 0.0632 -0.0727 0.0040 0.0061 -0.0046 0.0136
+ 0.0242 0.0085 0.0646 -0.0305 -0.1284 -0.0157 0.0376 0.0400 -0.0082 -0.0315
+ -0.0127 -0.0257 -0.0031 -0.0905 -0.0295 -0.0349 -0.0322 0.0358 -0.0159 0.1506
+ 0.0217 0.0263 -0.0179 -0.1047 0.0586 0.0046 -0.0901 -0.1336 0.0600 0.0547
+ 0.0320 0.0283 -0.0610 -0.0342 0.0462 -0.0214 -0.0862 0.0564 -0.0431 -0.1056
+ -0.0088 -0.0098 -0.0297 0.0749 0.0038 -0.1179 -0.0167 -0.0747 0.0329 0.0306
+ 0.0362 0.0625 -0.0019 -0.0076 0.1143 0.0523 0.1102 0.0852 0.0774 0.0721
+ -0.1238 0.0062 -0.0096 -0.0049 -0.0004 -0.0242 -0.0735 -0.0698 -0.0309 -0.0119
+ 0.0149 0.0095 0.0074 -0.0678 0.0921 0.0852 0.0107 -0.0623 -0.1286 0.0349
+ -0.0938 -0.1044 0.1349 0.0413 0.0082 -0.0250 -0.0245 -0.0342 0.0838 0.0701
+ 0.0395 0.0340 -0.0191 -0.0528 -0.1232 0.0854 0.0010 -0.0568 -0.0859 -0.1369
+ -0.0363 -0.0466 0.0019 -0.0095 0.0356 0.0114 0.0578 -0.0174 0.0490 0.1313
+ 0.0637 0.0652 0.0134 0.0533 -0.0112 -0.0275 0.1647 0.0985 0.0509 0.0515
+ 0.0047 -0.0278 -0.0342 0.0555 0.0335 -0.0198 -0.0151 0.0347 -0.0187 -0.0872
+ -0.0749 -0.0976 0.1020 0.0244 -0.0239 0.0342 -0.0172 -0.0524 -0.0990 -0.0836
+ -0.0501 -0.0766 0.0102 -0.0453 -0.0430 -0.0717 -0.0913 -0.1572 0.0174 0.0417
+ -0.0270 -0.0435 -0.0829 -0.0845 -0.1285 0.0375 -0.0105 -0.0635 -0.0457 -0.0855
+ -0.0255 -0.0537 -0.0065 0.0748 -0.0545 -0.1221 -0.0740 0.0503 -0.0141 -0.0440
+ 0.0197 0.0293 -0.0282 0.0881 0.0539 0.0014 -0.0512 -0.1213 -0.0122 -0.0763
+ 0.0155 0.0290 -0.0004 0.0346 0.0005 -0.0512 -0.0889 -0.1777 -0.0862 0.0577
+ 0.0492 0.0526 0.0295 0.0002 0.0717 0.0397 -0.0717 -0.0194 -0.0072 -0.0550
+ 0.0164 0.0268 -0.0641 -0.1311 -0.0675 0.0505 0.0258 0.0189 0.0475 0.0397
+ 0.0175 0.0087 0.0465 0.0745 0.0955 0.0324 0.0390 0.0004 0.0062 -0.0504
+ -0.0204 -0.0244 -0.0380 -0.0494 -0.0846 -0.1212 -0.0376 -0.1026 -0.1728 0.0703
+ 0.0060 -0.0144 0.0579 -0.0025 -0.0519 -0.0835 -0.0028 -0.0179 -0.0986 -0.1509
+ 0.0040 -0.0133 0.0153 -0.0236 -0.0845 -0.1477 0.1582 0.1264 0.0641 0.0506
+ -0.0352 -0.0554 -0.0777 -0.1243 0.0530 0.0317 -0.0428 -0.0616 -0.1301 0.0143
+ 0.0122 -0.0085 -0.0046 0.1390 0.0363 0.1067 0.0139 -0.0279 0.0389 -0.0089
+ 0.0142 -0.0023 0.0543 -0.0635 -0.0485 0.0599 -0.0472 0.0431 0.0222 -0.0028
+ -0.0221 -0.0361 -0.0559 -0.0935 -0.0724 -0.1297 0.0263 0.0058 -0.0343 -0.0902
+ 0.0137 0.0296 -0.0198 0.0044 -0.0718 -0.1038 0.0396 -0.0518 -0.0526 0.1067
+ 0.0751 0.0630 0.0749 0.0084 -0.0620 -0.0808 0.0475 -0.0005 0.0609 0.0380
+ -0.0316 0.0807 0.0816 0.0612 0.0295 -0.0039 -0.0705 -0.0695 0.0152 0.0050
+ 0.0096 0.0134 -0.0208 -0.0572 -0.1158 -0.1724 0.0643 0.0100 -0.0694 -0.0647
+ 0.0617 0.0717 0.0432 -0.0063 -0.0210 -0.0552 -0.0532 0.0478 -0.0396 -0.0843
+ -0.0085 -0.0240 -0.1077 0.0824 0.0348 -0.0217 0.0533 0.0001 0.0154 -0.0044
+ 0.0373 0.0120 0.1159 0.0733 0.0162 0.1225 0.0383 0.0285 0.0093 -0.0376
+ -0.0381 -0.0694 0.0125 -0.0220 -0.0212 0.0088 -0.0476 0.0571 0.1549 0.1049
+ -0.0101 -0.0181 -0.0630 -0.1343 -0.0779 -0.0981 0.1392 0.1052 0.0500 0.0369
+ 0.0108 0.0026 -0.0609 0.0817 0.0789 0.0266 -0.1107 0.0307 0.0982 0.0332
+ -0.0118 -0.0253 -0.0723 0.0773 0.0071 -0.0670 0.0489 -0.0164 -0.1178 -0.1125
+ 0.0195 0.0089 0.0084 -0.0338 -0.0800 0.1208 0.1946 0.1314 0.0501 0.0125
+ 0.0086 0.0007 -0.0207 0.0667 0.0856 0.0300 -0.0296 -0.0914 -0.1686 0.0065
+ -0.0144 -0.0199 -0.0912 0.0094 -0.0621 -0.0558 -0.1192 0.0172 0.1005 0.0459
+ 0.0097 -0.0154 0.0627 0.0100 0.0105 -0.0324 -0.1674 0.0110 0.0824 0.0256
+ -0.0049 0.0101 0.0343 0.0281 0.0643 -0.0030 0.0838 0.1121 0.0124 0.0561
+ 0.0337 0.0512 -0.0225 -0.0871 0.0212 0.1298 0.0677 0.0342 -0.0068 -0.0339
+ 0.0502 0.0655 0.0351 -0.0443 -0.0720 -0.0619 0.1007 0.1348 0.0552 0.0187
+ 0.0316 0.0371 -0.0035 0.0578 -0.0372 0.1005 0.0670 -0.0176 -0.0117 -0.0662
+ 0.0258 0.0044 0.1166 0.0904 0.0584 0.0033 -0.0458 0.1000 -0.0061 -0.0204
+ 0.0064 0.0162 -0.0308 -0.1166 0.0513 -0.0087 0.0658 -0.0118 0.1212 0.0973
+ 0.0186 0.0343 -0.0433 -0.0851 0.0491 -0.0004 -0.0221 0.0576 -0.0538 0.0712
+ -0.0343 -0.0464 -0.1078 -0.0036 0.0912 0.0273 0.0065 -0.0147 -0.0646 -0.0941
+ -0.0085 -0.0242 -0.0204 -0.0645 -0.0353 -0.0870 -0.2034 0.1233 0.0411 -0.0043
+ -0.0170 -0.0171 -0.0598 -0.0571 0.0580 -0.0064 -0.0462 -0.1078 -0.0861 0.1577
+ -0.0054 0.0012 -0.0199 -0.0119 0.0150 0.0024 0.1201 0.0220 0.0476 -0.0023
+ 0.0042 0.0022 -0.0222 -0.0180 -0.0727 -0.0994 -0.2163 -0.2307 0.0950 0.0551f
-10 128
- 0.0331 0.0140 0.0238 -0.0443 0.0011 -0.0332 0.0261 0.0967 0.0174 -0.0274
- 0.0004 -0.0159 -0.0151 -0.0035 -0.0676 0.0820 0.0239 -0.0128 -0.0534 -0.0913
- 0.0361 0.0088 0.0455 0.0643 0.0334 -0.0002 0.0634 0.0324 -0.0234 -0.0344
- -0.0419 0.0479 0.0338 0.0046 -0.0233 0.0848 0.0413 -0.0095 0.0078 -0.0078
- -0.0627 0.0222 0.0686 0.0179 -0.0321 -0.0445 -0.0409 0.0025 0.0127 0.0238
- 0.0105 -0.0122 0.0515 -0.0030 -0.0279 -0.0055 -0.0396 -0.0979 0.0637 0.0206
- -0.0443 -0.0094 0.0313 -0.0177 0.0073 -0.0035 -0.0126 -0.0421 -0.0754 -0.0631
- -0.0014 -0.0035 -0.0600 -0.0123 -0.0528 0.0430 0.0834 0.0469 0.0176 -0.0178
- -0.0084 -0.0075 -0.0323 0.0542 -0.0085 0.0081 -0.0140 0.0609 0.0950 0.0529
- 0.0038 -0.0313 0.0384 -0.0110 0.0706 0.0204 -0.0361 -0.0587 0.0082 -0.0450
- -0.0226 -0.0340 -0.0078 -0.0071 -0.0199 -0.0564 -0.0187 0.0098 -0.0425 0.1111
- -0.0051 -0.0223 -0.0021 -0.0237 0.0125 0.0308 0.1034 0.0819 0.0415 0.0180
- 0.0144 0.0316 0.0023 0.0180 -0.1024 0.0074 -0.0180 -0.0135 0.0122 -0.0202
- 0.0072 0.0557 -0.0270 -0.0570 0.0000 0.0482 0.0247 0.0367 0.0489 0.0253
- 0.0160 0.0037 0.0107 -0.0391 -0.0538 0.0422 -0.0610 0.0678 0.0721 0.0238
- 0.0002 -0.0165 0.0200 -0.0057 0.0260 0.0421 -0.0635 -0.0297 0.0236 0.0755
- 0.0029 0.0097 -0.0636 -0.0513 0.0052 -0.0202 -0.0135 -0.0569 -0.0044 -0.0437
- 0.0202 0.0254 0.0096 0.0227 0.0581 -0.0098 0.0725 0.0566 0.0566 0.0122
- 0.0056 -0.0078 0.0027 0.0077 -0.0370 0.0821 0.0069 0.1057 0.0056 -0.0204
- -0.0013 -0.0030 -0.0401 -0.0637 -0.0196 -0.0747 0.0805 0.0440 -0.0118 -0.0082
- 0.0212 0.0244 -0.0090 0.0512 0.0310 0.0859 0.0558 0.0357 0.0148 -0.0117
- 0.0394 -0.0050 -0.0605 -0.0214 -0.0334 0.0118 -0.0121 0.0166 0.0220 -0.0191
- 0.0527 0.0281 0.0379 -0.0096 -0.0464 0.0565 0.0584 0.0217 0.0176 -0.0150
- 0.0044 -0.0019 -0.0442 0.0239 0.0005 -0.0295 -0.0319 -0.0916 0.0052 0.0843
- -0.0072 -0.0260 -0.0009 -0.0607 -0.0215 0.0909 0.0301 0.0194 -0.0389 0.0372
- -0.0244 0.0655 0.0381 -0.0053 0.0323 -0.0132 -0.0622 0.0348 -0.0268 0.0414
- 0.0282 0.0127 0.0642 0.0200 -0.0010 -0.0178 0.0428 -0.0176 0.0523 0.0583
- -0.0222 -0.0522 0.0121 -0.0280 0.0097 0.0083 0.0693 0.0216 -0.0331 -0.0679
- 0.0040 -0.0051 -0.0054 0.0593 0.0694 0.0291 0.0212 -0.0131 -0.0355 -0.0828
- -0.0114 -0.0214 -0.0291 -0.0100 0.0484 0.0514 -0.0050 -0.0481 -0.0928 -0.0382
- 0.0043 -0.0232 -0.0107 0.0254 -0.0529 -0.0725 0.0069 -0.0543 -0.0307 -0.0074
- -0.0308 -0.0429 -0.0494 -0.0547 -0.0367 -0.0303 0.0010 0.0006 0.0436 -0.0033
- -0.0135 -0.0265 -0.0663 0.0275 -0.0292 0.0621 0.0054 -0.0280 -0.0684 0.0520
- -0.0220 -0.0339 -0.0628 0.0426 0.0043 -0.0515 0.0551 0.0182 0.0298 0.0200
- -0.0161 -0.0366 -0.0145 -0.0313 -0.0542 0.0531 -0.0287 -0.0835 0.0295 -0.0025
- 0.0136 0.0365 -0.0070 -0.0006 -0.0034 -0.0537 -0.0824 -0.0300 -0.0443 -0.0537
- -0.0412 -0.0461 0.0280 -0.0295 0.0462 0.0160 0.0182 -0.0363 -0.0295 0.0578
- 0.0067 -0.0079 0.0136 0.0280 -0.0380 -0.0697 0.0974 0.0693 0.0221 -0.0083
- 0.0644 0.0386 0.0082 0.0379 0.0217 0.0044 -0.0131 -0.0544 -0.0200 -0.0410
- 0.0075 0.0235 -0.0184 -0.0241 0.0992 0.0521 -0.0029 -0.0460 0.0054 0.0294
- 0.0302 0.0155 -0.0324 0.0561 -0.0174 -0.0466 0.0500 -0.0231 0.0523 -0.0057
- 0.0189 -0.0051 0.0323 0.0721 0.0506 0.0319 0.0078 -0.0301 0.0647 0.0208
- 0.0006 -0.0303 0.0657 -0.0290 -0.0723 -0.0058 -0.0059 0.0045 0.0247 -0.0050
- 0.0027 -0.0013 -0.0206 -0.0098 0.0378 -0.0088 -0.1201 0.0004 0.0469 -0.0132
- -0.0096 0.0042 -0.0202 -0.0162 -0.0707 -0.0748 0.0366 -0.0244 0.0670 0.0343
- 0.0122 0.0133 -0.0012 -0.0725 0.0284 0.0541 0.0377 0.0012 -0.0471 -0.0657
- 0.0054 -0.0050 -0.0173 -0.0643 -0.0897 0.0083 0.0317 -0.0231 0.0189 -0.0349
- 0.0479 0.0364 0.0175 -0.0047 0.0490 0.0273 -0.0524 0.0496 0.0271 0.0221
- -0.0057 -0.0468 0.0174 0.0310 0.0396 -0.0174 0.0071 0.0750 0.0193 -0.0407
- -0.0165 -0.0238 -0.0145 0.0325 0.0240 -0.0095 0.0837 0.0278 -0.0629 0.0425
- -0.0013 0.0106 -0.0138 -0.0302 0.0124 -0.0208 -0.0322 0.0570 -0.0919 0.0108
- 0.0250 -0.0036 0.0778 0.0282 -0.0181 0.0658 -0.0151 -0.0289 0.0024 -0.0424
- 0.0052 -0.0168 0.0502 0.0163 0.0336 -0.0055 -0.0014 0.0747 -0.0148 0.0879
- -0.0007 -0.0380 0.0358 -0.0341 -0.0016 -0.0841 0.0500 0.0108 0.0258 0.0098
- -0.0103 -0.0218 -0.0027 0.0836 0.0311 0.0410 -0.0658 0.0210 -0.0035 -0.0311
- 0.0036 -0.0205 0.0554 0.0110 -0.0399 -0.0077 0.0540 0.0158 -0.0889 -0.0191
- -0.0093 -0.0217 -0.0559 -0.0323 0.0064 -0.0082 -0.0505 0.1094 0.0233 0.0104
- -0.0027 -0.0063 -0.0228 0.0319 0.0039 -0.0035 -0.0524 -0.0983 -0.0196 -0.0816
- 0.0007 -0.0054 -0.0087 0.0718 0.0673 0.0261 0.0003 -0.0382 -0.0487 0.0642
- 0.0414 0.0311 -0.0184 0.0372 -0.0070 -0.0489 0.0336 0.0045 -0.0505 0.0646
- -0.0016 -0.0326 0.0291 -0.0384 0.0080 0.0194 -0.0773 0.0538 -0.0259 -0.0623
- 0.0042 0.0057 0.0095 -0.0028 -0.0474 -0.0106 -0.0336 -0.0917 -0.1022 0.0406
- -0.0136 -0.0185 -0.0724 0.0194 -0.0479 -0.0140 -0.0757 0.0021 0.0269 0.0069
- -0.0329 -0.0622 -0.0409 -0.0050 0.0310 -0.0029 -0.0395 -0.0014 -0.0338 -0.0037
- 0.0021 0.0356 -0.0038 -0.0436 0.0328 -0.0135 0.0642 0.0082 -0.0366 0.0725
- 0.0032 -0.0226 0.0198 0.0280 -0.0233 0.0797 0.0590 -0.0209 -0.0149 0.0414
- 0.0142 0.0292 -0.0210 0.0102 0.0193 -0.0114 0.0858 0.0370 -0.0288 -0.0721
- -0.0300 0.0511 0.0014 -0.0520 -0.0293 -0.0280 -0.0316 -0.0618 0.0265 0.0022
- -0.0466 -0.0384 0.0281 -0.0192 -0.0173 -0.0394 -0.0151 0.0920 0.0264 0.0134
- 0.0104 0.0448 -0.0081 0.0134 -0.0196 -0.0696 -0.0251 0.0656 0.0342 0.0038
- 0.0085 0.0002 -0.0049 -0.0119 -0.0582 0.0666 0.0333 -0.0120 0.0943 0.0581
- -0.0053 -0.0194 -0.0322 0.0674 0.0108 -0.0630 -0.0583 0.0332 -0.0311 -0.0008
- 0.0009 -0.0191 -0.0084 -0.0479 0.0630 0.0152 0.0364 -0.0387 0.0873 0.0333
- -0.0284 -0.0480 0.0674 0.0594 0.0240 0.0460 0.0357 0.0119 0.0218 0.0001
- 0.0047 0.0330 0.0242 -0.0157 -0.0531 -0.0897 -0.0024 -0.0026 -0.0470 0.0392
- 0.0141 0.0081 -0.0390 0.0135 -0.0034 0.0998 -0.0108 -0.0435 0.0396 -0.0056
- -0.0253 0.0484 0.0575 0.0493 0.0447 0.0176 -0.0247 -0.0445 -0.0268 -0.0338
- 0.0221 0.0088 -0.0394 0.0439 0.0210 -0.0151 -0.0080 -0.0381 -0.1057 -0.0241
- 0.0045 -0.0142 -0.0148 -0.1049 0.0239 -0.0108 -0.0227 0.0356 0.0298 0.0291
- 0.0259 0.0138 0.0545 0.0399 0.0119 -0.0288 -0.0525 0.0369 -0.0402 -0.0714
- 0.0126 0.0019 0.0168 0.0132 -0.0403 -0.0322 -0.1177 0.0374 -0.0328 0.0688
- -0.0122 -0.0087 -0.0245 -0.0224 -0.0416 -0.0666 0.0246 -0.0127 -0.0490 -0.1004
- 0.0380 0.0432 -0.0029 -0.0318 -0.0515 -0.0173 0.0218 -0.0105 -0.0756 -0.0406
- -0.0303 -0.0502 -0.0137 0.0117 -0.0753 0.0318 0.0257 0.0355 0.0225 0.0047
- -0.0004 -0.0090 -0.0123 -0.0169 -0.0579 -0.0569 -0.1260 -0.1069 0.0203 0.0183
- -0.0513 0.0306 0.0291 0.0125 0.0133 -0.0132 0.0575 0.0286 0.0718 0.0632
- 0.0159 0.0159 0.0684 0.0424 -0.0104 0.0235 -0.0109 0.0793 0.0514 0.0103
- -0.0189 -0.0252 -0.0397 -0.0579 0.0360 0.0704 0.0063 0.0114 0.0053 -0.0474
- 0.0154 0.0019 -0.0634 -0.0005 0.0707 0.0161 0.0400 0.0428 -0.0240 0.0053
- 0.0463 0.0411 0.0044 -0.0333 -0.0399 -0.0424 0.0288 0.0026 0.0574 0.0349
- -0.0440 0.0233 0.0411 0.0227 0.0604 0.0236 0.0455 0.0564 -0.0089 -0.0337
- 0.0200 -0.0102 0.0516 -0.0264 0.0798 0.0610 0.0461 0.0285 0.0084 -0.0128
- 0.0056 -0.0197 0.0184 0.0185 -0.0115 0.0258 0.0488 -0.0095 0.0378 -0.0467
- -0.0160 -0.0275 -0.0630 0.0411 0.0026 0.0008 0.0246 0.0143 -0.0397 -0.0783
- 0.0287 0.0103 0.0180 -0.0400 -0.0527 0.0302 -0.0431 -0.0674 -0.0198 -0.0667
- 0.0248 0.0268 -0.0210 0.0160 -0.0054 0.0438 -0.0391 0.0117 -0.0422 -0.0909
- -0.0109 -0.0259 -0.0233 -0.0397 -0.0817 0.0038 -0.0649 0.0017 -0.0659 -0.0197
- -0.0060 -0.0036 -0.0610 0.0391 0.0603 0.0055 -0.0092 -0.0487 0.0421 -0.0159
- -0.0781 -0.0043 -0.0051 -0.0053 0.0352 0.0059 -0.0401 -0.0178 0.0339 0.0048
- -0.0169 -0.0240 -0.0445 -0.0413 -0.0030 -0.0487 0.0243 -0.0333 -0.0934 0.0163
- -0.0025 -0.0043 0.0228 0.1022 -0.0224 -0.0228 0.0191 -0.0046 -0.0043 -0.0059
- 0.0281 0.0043 0.0595 0.0120 0.0109 -0.0540 -0.0743 0.0183 0.0529 0.0063
- -0.0169 -0.0375 0.0310 0.0318 -0.0331 -0.0330 -0.0762 -0.0173 -0.0055 -0.0571
- 0.0289 0.0161 -0.0396 0.0891 0.0374 0.0106 -0.0196 0.0614 0.0195 -0.0065
- 0.0363 0.0099 0.0535 0.0015 -0.0277 -0.0609 0.0244 -0.0320 -0.0105 -0.0614
- 0.0015 -0.0019 -0.0185 -0.0495 -0.0772 -0.0050 0.0227 0.0075 -0.0235 0.0901
- -0.0084 -0.0314 -0.0123 -0.0590 -0.0031 -0.0622 -0.0393 -0.0799 0.0214 -0.0346
- -0.0264 -0.0235 -0.0640 -0.0518 0.0392 0.0064 0.0438 0.0012 0.0051 0.0452
- 0.0102 0.0068 -0.0395 0.0338 0.0296 -0.0042 0.0643 0.0102 0.0559 0.0928
- -0.0062 -0.0125 -0.0097 -0.0151 0.0683 0.0582 -0.0037 0.0674 0.0678 0.0394
- -0.0023 -0.0097 -0.0221 -0.0496 0.0712 -0.0216 -0.0130 0.0238 -0.0221 -0.0721
- 0.0040 -0.0016 -0.0189 -0.0486 -0.0467 -0.0745 -0.0861 0.0379 0.0106 -0.0242
- 0.0065 -0.0204 0.0610 0.0462 0.0154 -0.0237 -0.0338 -0.0616 -0.0642 0.0092
- 0.0214 0.0106 0.0286 -0.0028 0.0757 0.0130 0.0291 -0.0013 -0.0889 -0.0069
- -0.0097 0.0663 0.0169 0.0403 0.0104 0.0135 -0.0328 -0.0516 0.0770 0.0410
- 0.0324 0.0094 0.0360 -0.0668 0.0072 -0.0257 -0.0108 -0.0538 -0.0412 0.0247
- 0.0222 0.0352 0.0225 0.0112 -0.0068 0.0642 0.0194 -0.0211 -0.0967 0.0409
- -0.0113 -0.0479 0.0240 -0.0150 -0.0060 -0.0112 -0.0075 0.0175 0.1012 0.0637
- -0.0185 0.0289 -0.0059 0.0278 0.0226 -0.0652 0.0168 -0.0469 -0.0129 -0.0059
- 0.0163 0.0887 0.0372 0.0273 0.0081 -0.0167 0.0280 0.0110 -0.0139 -0.0224
- -0.0040 -0.0014 -0.0326 -0.0406 0.0036 -0.0520 -0.0185 -0.0254 0.0988 0.0697
- -0.0874 0.0724 -0.0256 0.0227 0.0017 -0.0065 -0.0133 -0.0286 -0.0499 -0.0222
- -0.0037 0.0041 -0.0325 -0.0603 0.0295 -0.0020 -0.0668 -0.0552 -0.0643 0.0571
- 0.0427 0.0266 -0.0122 0.0151 -0.0276 -0.0337 -0.0581 -0.0633 0.0618 0.0055
- 0.0178 0.0464 0.0242 -0.0288 0.0417 0.0194 -0.0065 -0.0173 0.0200 -0.0409
- -0.0309 -0.0531 -0.0092 0.0458 0.0012 0.0079 -0.0126 -0.0604 0.0546 0.0273
- -0.0304 0.0461 0.0475 -0.0405 -0.0342 -0.0273 0.0569 0.0392 -0.0095 -0.0377
- -0.0211 -0.0380 -0.0398 0.0478 0.0498 0.0571 0.0397 0.0191 0.0284 0.0027
+10 256
+ -0.0448 0.0001 0.0172 0.0154 0.0284 0.0165 0.0049 0.0650 -0.0157 -0.0570
+ -0.0027 -0.0099 -0.0142 -0.0054 0.0428 0.0157 -0.0946 -0.0007 -0.0573 -0.0002
+ -0.0257 -0.0126 -0.0394 0.0355 0.0016 -0.0372 0.0659 0.0098 -0.0417 0.0339
+ 0.0120 0.0079 -0.0497 0.0218 0.0193 -0.0455 -0.0286 -0.0154 0.0650 0.0313
+ -0.0112 -0.0132 -0.0429 -0.0882 0.0391 -0.0064 -0.0198 0.0007 -0.0053 -0.0190
+ -0.0376 0.0348 0.0048 -0.0455 0.0447 0.0044 0.0421 0.0439 -0.0146 0.0212
+ -0.0220 0.0234 -0.0195 -0.0506 0.0390 0.0287 0.0162 -0.0253 -0.0285 -0.0719
+ -0.0078 0.0123 0.0116 -0.0048 -0.0276 -0.0431 0.0057 0.0633 -0.0132 0.0985
+ 0.0371 0.0052 0.0539 0.0328 0.0008 -0.0286 0.0567 -0.0176 0.0381 -0.0048
+ -0.0026 -0.0188 -0.0292 0.0289 -0.0405 -0.0465 -0.0549 0.0258 0.0262 0.0591
+ 0.0415 0.0234 0.0114 0.0329 -0.0190 0.0757 0.0361 0.0241 0.0332 0.0085
+ 0.0088 0.0066 -0.0519 0.0015 0.0389 -0.0339 0.0588 0.0153 -0.0093 -0.0500
+ -0.0305 -0.0453 0.0247 0.0006 -0.0163 0.0617 -0.0091 0.0238 0.0469 0.0053
+ -0.0265 0.0127 0.0480 -0.0052 -0.0661 -0.0004 -0.0418 0.0376 -0.0022 0.0040
+ 0.0207 0.0042 0.0211 -0.0308 0.0263 -0.0237 -0.0963 0.0393 -0.0146 -0.0574
+ 0.0077 0.0134 -0.0217 0.0058 -0.0260 -0.0553 -0.0193 -0.1024 0.0184 -0.0067
+ -0.0021 -0.0218 -0.0028 -0.0411 -0.0440 -0.0191 -0.0038 0.0653 -0.0654 0.0175
+ -0.0912 0.0358 0.0150 -0.0160 -0.0189 0.0112 0.0609 0.0264 -0.0082 -0.0228
+ -0.0137 -0.0348 0.0099 0.0875 0.0229 0.0329 -0.0068 -0.0321 0.0131 -0.0267
+ 0.0021 -0.0057 -0.0108 -0.0337 -0.0870 0.0699 0.0162 -0.0003 -0.0213 -0.0526
+ 0.0095 0.0212 -0.0131 0.0339 0.0486 0.0083 -0.0527 -0.0828 -0.0263 -0.0044
+ 0.0100 -0.0115 0.0105 0.0016 0.0703 -0.0063 -0.0520 0.0452 0.0468 0.0136
+ 0.0031 0.0679 0.0498 0.0112 0.0089 -0.0109 -0.0175 -0.0306 0.0454 0.0190
+ 0.0140 0.0033 0.0250 0.0414 0.0302 0.0656 0.0106 -0.0438 0.0425 0.0297
+ -0.0136 -0.0089 0.0292 -0.0570 -0.0207 -0.0578 -0.0200 0.0096 -0.0288 -0.0290
+ -0.0237 -0.0340 -0.0348 -0.0200 0.0531 0.0455 -0.0336 0.0156 0.0101 0.0120
+ 0.0330 0.0327 0.0085 -0.0142 0.0080 0.0400 -0.0713 0.0304 0.0189 -0.0283
+ -0.0056 -0.0148 -0.0302 -0.0493 -0.0305 -0.0071 -0.0171 -0.0544 0.0707 0.0573
+ -0.0000 0.0336 0.0128 -0.0021 0.0394 0.0500 -0.0019 0.0707 0.0548 0.0175
+ 0.0140 0.0079 -0.0015 -0.0309 0.0439 -0.0548 0.0122 -0.0451 0.0653 0.0012
+ -0.0019 -0.0209 0.0433 -0.0081 0.0258 0.0271 0.0618 -0.0088 0.0476 -0.0092
+ -0.0156 -0.0215 -0.0410 -0.0159 -0.0253 -0.0529 -0.0907 0.0492 -0.0023 -0.0309
+ 0.0139 0.0095 0.0349 0.0026 0.0672 -0.0409 0.0208 0.0123 -0.0219 -0.0074
+ 0.0400 0.0187 0.0627 0.0243 0.0461 0.0320 -0.0099 0.0219 0.0223 -0.0110
+ 0.0353 -0.0009 -0.0143 -0.0083 0.0480 0.0310 0.0143 0.0290 0.0228 0.0766
+ -0.0108 -0.0354 -0.0365 0.0355 0.0450 -0.0147 -0.0249 0.0253 -0.0333 -0.0520
+ -0.0123 -0.0234 -0.0603 -0.0080 0.0498 -0.0231 0.0378 -0.0064 0.0153 0.0342
+ -0.0098 -0.0323 0.0564 0.0125 -0.0418 -0.0654 0.0277 0.0060 -0.0142 -0.0282
+ -0.0191 -0.0472 0.0006 -0.0377 0.0193 -0.0421 -0.0317 -0.0306 0.0425 -0.0281
+ 0.0027 -0.0125 0.0462 0.0635 0.0140 -0.0097 -0.0839 0.0006 -0.0341 0.0384
+ 0.0638 0.0314 -0.0167 0.0140 0.0315 -0.0149 0.0075 -0.0323 0.0169 -0.0043
+ -0.0250 -0.0253 -0.0164 0.0271 0.0231 0.0844 0.0328 0.0004 -0.0434 -0.0305
+ -0.0174 -0.0299 0.0067 0.0013 -0.0381 0.0698 0.0693 0.0229 -0.0497 0.0203
+ 0.0268 0.0024 -0.0281 -0.0442 0.0609 -0.0170 -0.0262 -0.0295 -0.0590 0.0296
+ 0.0019 0.0100 0.0063 0.0563 -0.0047 -0.0521 0.0342 -0.0262 -0.0250 -0.0731
+ -0.0019 0.0083 -0.0167 -0.0250 0.0905 0.0295 -0.0235 0.0240 -0.0085 -0.0425
+ 0.0334 0.0360 0.0001 0.0473 0.0096 -0.0164 -0.0682 0.0085 -0.0325 0.0089
+ 0.0292 0.0194 0.0188 -0.0164 0.0557 0.0163 -0.0091 -0.0390 -0.0111 -0.0739
+ -0.0177 -0.0293 -0.0252 0.0204 -0.0248 -0.0599 0.0085 -0.0476 0.0669 0.0233
+ -0.0207 0.0434 0.0423 -0.0150 -0.0257 -0.0481 -0.0376 -0.0145 -0.0025 -0.0584
+ 0.0276 -0.0108 -0.0167 0.0067 0.0384 0.0024 -0.0631 -0.0634 0.0484 0.0063
+ 0.0410 0.0029 0.0598 -0.0185 -0.0220 0.0256 0.0253 0.0574 -0.0088 0.0092
+ 0.0108 -0.0016 -0.0568 -0.0072 -0.0287 0.0855 0.0168 0.0099 0.0098 0.0242
+ 0.0275 0.0025 0.0123 -0.0447 -0.0600 0.0458 -0.0256 0.0106 0.0509 0.0257
+ 0.0057 -0.0424 -0.0102 0.0254 -0.0156 -0.0447 0.0341 -0.0178 0.0159 -0.0286
+ 0.0157 -0.0038 0.0302 0.0133 0.0647 0.0414 0.0291 0.0311 -0.0570 -0.0204
+ -0.0420 -0.0500 -0.0247 -0.0427 -0.0200 -0.0130 -0.0107 0.0257 0.0187 0.0024
+ -0.0010 0.0157 -0.0007 0.0712 0.0222 0.0342 0.0283 -0.0049 -0.0583 0.0319
+ 0.0326 0.0282 0.0329 -0.0250 -0.0054 -0.0024 -0.0637 0.0180 -0.0274 0.0587
+ 0.0188 0.0090 -0.0607 0.0733 -0.0295 0.0117 0.0048 -0.0001 -0.0214 -0.0250
+ 0.0306 0.0675 -0.0083 -0.0094 0.0439 0.0222 0.0156 0.0079 0.0135 0.0124
+ -0.0203 -0.0449 0.0442 0.0304 0.0079 -0.0088 -0.0630 0.0592 0.0252 -0.0248
+ -0.0540 -0.0016 -0.0415 0.0400 0.0277 0.0061 -0.0010 -0.0363 -0.0122 -0.0326
+ 0.0085 -0.0064 -0.0343 -0.0197 -0.0755 0.0224 -0.0976 0.0249 -0.0120 -0.0033
+ 0.0192 -0.0045 -0.0124 -0.0214 -0.0409 0.0415 0.0901 0.0557 0.0273 -0.0086
+ -0.0035 -0.0273 0.0433 0.0416 0.0390 -0.0103 0.0411 -0.0178 -0.0205 0.0723
+ 0.0248 0.0158 -0.0189 -0.0117 -0.0810 -0.0115 0.0471 0.0118 -0.0131 0.0420
+ 0.0265 0.0408 0.0017 -0.0609 0.0058 -0.0470 0.0231 0.0159 0.0247 0.0172
+ 0.0560 0.0283 0.0232 0.0313 -0.0039 -0.0030 -0.0039 -0.0184 -0.0273 -0.0637
+ 0.0319 -0.0097 0.0658 -0.0493 0.0200 -0.0168 0.0159 -0.0140 0.0371 -0.0051
+ -0.0129 -0.0201 -0.0616 -0.0031 -0.0652 0.0124 0.0240 -0.0333 0.0030 -0.0068
+ -0.0164 -0.0268 -0.0498 -0.0044 0.0318 -0.0183 -0.0477 -0.0689 0.0124 -0.0398
+ -0.0674 0.0499 0.0285 -0.0035 0.0284 0.0140 -0.0336 -0.0695 -0.0080 -0.0162
+ 0.0514 0.0225 0.0104 -0.0118 -0.0291 -0.0267 0.0626 0.0085 -0.0026 -0.0504
+ -0.0289 -0.0626 0.0145 0.0294 -0.0081 -0.0195 -0.0177 -0.0147 -0.0665 -0.0081
+ 0.0323 0.0014 -0.0027 -0.0791 0.0374 0.0404 0.0081 -0.0446 0.0359 0.0011
+ -0.0436 -0.0039 0.0279 -0.0073 -0.0466 0.0171 0.0032 -0.0564 0.0110 -0.0309
+ 0.0014 -0.0049 -0.0169 -0.0744 -0.0495 -0.0109 0.0483 -0.0024 0.0018 -0.0113
+ 0.0096 0.0230 -0.0234 0.0300 0.0395 0.0028 0.0213 -0.0088 -0.0767 -0.0434
+ 0.0161 -0.0040 -0.0063 0.0754 0.0118 0.0056 0.0213 0.0662 -0.0067 -0.0473
+ 0.0073 0.0108 0.0361 -0.0158 -0.0203 0.0971 0.0182 -0.0265 0.0153 0.0004
+ 0.0133 -0.0123 0.0149 -0.0131 -0.0651 0.0162 -0.0193 0.0851 -0.0064 -0.0599
+ 0.0052 0.0067 -0.0245 0.0232 -0.0467 -0.0713 -0.0036 0.0355 0.0285 -0.0319
+ 0.0072 0.0180 -0.0099 -0.0709 -0.0310 0.0024 -0.0282 0.0502 0.0075 -0.0490
+ -0.0082 -0.0056 -0.0460 -0.0437 -0.0132 -0.0471 0.0513 -0.0031 0.0648 0.0065
+ -0.0056 -0.0269 0.0377 -0.0017 0.0200 0.0260 -0.0426 -0.0225 0.0924 0.0452
+ 0.0022 0.0022 -0.0441 0.0285 0.0551 0.0354 0.0408 0.0650 -0.0064 0.0016
+ 0.0009 -0.0366 -0.0257 0.0224 0.0017 0.0466 0.0064 -0.0264 -0.0167 -0.0743
+ -0.0401 0.0678 0.0055 0.0270 0.0305 -0.0204 -0.0651 0.0379 -0.0062 0.0080
+ -0.0115 -0.0488 0.0131 -0.0308 0.0014 -0.0381 0.0385 -0.0273 0.0224 0.0537
+ 0.0198 0.0236 -0.0173 0.0366 0.0517 0.0692 0.0156 -0.0144 -0.0002 -0.0398
+ 0.0110 0.0119 -0.0118 -0.0136 0.0295 0.0079 0.0975 0.0227 0.0070 0.0447
+ 0.0110 -0.0127 -0.0045 0.0650 0.0163 -0.0349 -0.0070 -0.0545 -0.0516 0.0091
+ -0.0071 -0.0150 -0.0348 -0.0263 0.0004 -0.0201 0.0495 0.0650 -0.0152 -0.0758
+ -0.0573 0.0169 0.0779 0.0164 0.0250 -0.0273 0.0231 0.0053 -0.0041 -0.0173
+ 0.0089 0.0027 -0.0220 -0.0303 0.0726 0.0734 0.0155 -0.0220 -0.0385 0.0263
+ 0.0144 0.0021 0.0268 -0.0491 0.0216 0.0166 0.0428 -0.0283 -0.0654 0.0027
+ -0.0128 0.0148 0.0219 -0.0578 0.0282 0.0011 -0.0411 0.0167 0.0522 0.0075
+ -0.0060 0.0002 -0.0306 0.0240 0.0041 0.0400 0.0898 0.0142 0.0061 -0.0479
+ 0.0104 -0.0304 0.0280 -0.0214 0.0207 -0.0211 0.0092 0.0733 0.0070 -0.0557
+ 0.0122 0.0385 -0.0594 -0.0044 -0.0052 -0.0147 0.0277 0.0331 0.0079 0.0074
+ 0.0135 0.0007 0.0138 -0.0469 0.0263 0.0010 -0.0353 -0.0834 -0.0121 0.0849
+ 0.0107 0.0018 -0.0091 -0.0101 0.0069 -0.0245 0.0251 0.1183 0.0538 0.0334
+ -0.0043 -0.0108 -0.0031 0.0354 -0.0020 -0.0367 0.0879 0.0530 0.0470 -0.0025
+ 0.0114 -0.0162 0.0573 0.0295 0.0351 0.0044 0.0432 0.0581 0.0629 0.0306
+ -0.0373 -0.0460 -0.0033 -0.0126 0.0313 0.0058 0.0443 0.0003 -0.0093 -0.0583
+ 0.0139 -0.0175 0.0280 0.0193 -0.0351 -0.0457 -0.0758 0.0120 -0.0254 -0.0525
+ 0.0068 -0.0083 -0.0011 0.0322 0.0357 -0.0078 0.0281 -0.0148 0.0992 0.0421
+ -0.0325 0.0186 0.0412 -0.0516 -0.0331 -0.0274 0.0218 0.0014 0.0363 0.0486
+ 0.0049 0.0170 0.0039 0.0015 -0.0092 -0.0373 -0.1017 -0.0148 0.0606 0.0267
+ 0.0150 0.0027 0.0289 0.0252 0.0214 -0.0304 -0.0427 0.1018 -0.0115 0.0128
+ 0.0062 0.0444 0.0312 0.0058 -0.0155 -0.0412 -0.0221 -0.0629 -0.0553 -0.0003
+ 0.0005 -0.0425 0.0109 0.0455 -0.0150 0.0230 0.0186 -0.0164 0.0300 0.0524
+ 0.0289 0.0143 -0.0338 -0.0170 -0.0616 -0.0002 -0.0097 -0.0358 0.0131 -0.0453
+ -0.0042 0.0347 0.0366 0.0406 0.0316 0.0283 0.0771 0.0281 -0.0064 -0.0114
+ -0.0376 0.0209 -0.0031 -0.0180 0.0168 -0.0250 -0.0023 -0.0501 -0.0893 -0.0054
+ -0.0048 -0.0375 0.0135 -0.0205 -0.0718 -0.0003 0.0463 0.0024 0.0359 -0.0185
+ 0.0336 0.0044 -0.0568 -0.0200 0.0348 0.0240 -0.0225 0.0005 -0.0074 -0.0497
+ -0.0184 -0.0202 -0.0416 -0.0565 0.0202 0.0384 0.0588 0.0440 0.0042 -0.0034
+ 0.0197 -0.0064 0.0661 0.0062 0.0058 -0.0545 -0.0129 -0.0020 0.0062 0.0559
+ 0.0073 -0.0366 -0.0099 -0.0712 -0.0113 -0.0222 -0.0578 0.0281 0.0058 -0.0411
+ 0.0370 0.0419 0.0354 0.0095 -0.0026 -0.0056 0.0367 -0.0122 -0.0337 0.0501
+ -0.0110 -0.0208 -0.0219 -0.0435 -0.0736 0.0396 -0.0008 0.0658 0.0535 0.0217
+ 0.0099 -0.0010 -0.0004 0.0377 -0.0213 -0.0165 -0.0697 -0.0756 -0.0017 -0.0491
+ -0.0092 -0.0205 -0.0110 0.0257 0.0743 0.0343 0.0003 -0.0560 -0.0179 0.0010
+ 0.0212 -0.0231 -0.0076 -0.0360 -0.0386 0.0158 -0.0052 -0.0883 -0.0250 -0.0030
+ -0.0264 -0.0185 -0.0380 -0.0198 -0.0400 -0.0710 0.0407 -0.0112 -0.0120 -0.0587
+ 0.0533 0.0061 -0.0203 0.0316 -0.0173 -0.0140 -0.0364 0.0670 0.0168 -0.0195
+ 0.0269 0.0184 -0.0132 -0.0514 0.0285 0.0205 -0.0109 0.0852 -0.0219 0.0117
+ 0.0025 -0.0029 -0.0446 0.0714 0.0095 -0.0356 -0.0928 0.0183 -0.0195 -0.0024
+ -0.0033 0.0104 -0.0126 0.0083 -0.0147 0.0649 -0.0188 -0.0692 -0.0389 -0.0541
+ -0.0035 0.0503 0.0364 -0.0331 0.0282 0.0101 0.0018 0.0244 -0.0381 -0.0433
+ 0.0099 -0.0018 0.0365 -0.0126 -0.0485 -0.0076 0.0295 -0.0091 -0.0528 -0.0805
+ 0.0140 -0.0008 -0.0440 0.0647 0.0176 0.0160 0.0544 0.0137 0.0505 0.0373
+ -0.0174 -0.0283 -0.0353 -0.0398 -0.0626 0.0334 0.0001 -0.0219 -0.0686 0.0596
+ 0.0290 0.0104 0.0133 0.0771 0.0561 0.0145 -0.0056 -0.0142 0.0103 0.0076
+ -0.0209 -0.0402 -0.0385 0.0139 0.0048 0.0519 0.0129 -0.0451 0.0558 0.0192
+ 0.0162 -0.0129 0.0404 -0.0022 -0.0527 -0.0236 0.0147 -0.0347 -0.0649 0.0510
+ -0.0080 -0.0000 0.0193 0.0361 0.0420 0.0265 -0.0665 -0.0170 0.0091 -0.0466
+ -0.0045 -0.0174 -0.0112 -0.0405 -0.0285 -0.0756 0.0136 -0.0180 -0.0802 0.0317
+ 0.0027 -0.0033 0.0137 0.0030 0.0047 0.0192 -0.0287 -0.0689 -0.1230 0.0543
+ -0.0153 -0.0317 -0.0327 -0.0070 -0.0490 -0.0534 0.0550 0.0315 0.0044 0.0483
+ -0.0001 -0.0091 -0.0360 0.0283 -0.0340 -0.0101 -0.0276 -0.0497 -0.0943 -0.0275
+ -0.0432 0.0411 -0.0076 -0.0379 -0.0246 0.0780 -0.0335 0.0379 0.0241 0.0128
+ 0.0249 0.0394 0.0034 0.0102 0.0357 -0.0062 0.0338 0.0552 0.0067 -0.0545
+ 0.0020 -0.0217 -0.0179 -0.0182 -0.0009 0.0758 -0.0288 0.0762 0.0100 -0.0224
+ 0.0492 0.0250 -0.0089 -0.0391 -0.0138 -0.0266 -0.0202 -0.0399 -0.0537 -0.0067
+ -0.0377 0.0355 0.0496 0.0044 0.0038 -0.0020 0.0115 -0.0098 -0.0519 0.0716
+ -0.0043 0.0014 -0.0412 -0.0023 0.0181 -0.0058 -0.0509 0.0774 0.0545 -0.0183
+ 0.0151 0.0197 -0.0013 -0.0421 -0.0514 0.0459 -0.0029 -0.0106 -0.0887 0.0042
+ 0.0024 -0.0044 -0.0156 -0.0101 -0.0391 -0.0463 -0.1080 -0.0096 -0.0727 0.0632
+ 0.0033 -0.0211 0.0388 0.0430 -0.0070 0.0680 0.0433 0.0466 0.0099 -0.0310
+ -0.0225 -0.0464 -0.0129 0.0276 0.0172 0.0052 0.0324 0.0687 0.0261 0.0244
+ 0.0302 0.0094 0.0490 0.0171 0.0291 0.0185 -0.0277 -0.0541 -0.0558 -0.0120
+ -0.0001 0.0467 -0.0029 0.0710 0.0077 -0.0252 0.0136 -0.0027 0.0405 0.0033
+ 0.0002 0.0030 -0.0209 -0.0679 -0.0126 -0.0342 -0.0270 -0.0723 0.0277 -0.0355
+ -0.0534 -0.0090 0.0014 0.0227 -0.0068 0.0107 -0.0668 0.0277 -0.0353 0.0357
+ -0.0151 -0.0289 -0.0291 -0.0370 0.0230 -0.0005 -0.0091 -0.0502 -0.0793 -0.0466
+ -0.0030 -0.0251 0.0107 0.0607 0.0037 -0.0718 0.0005 0.0495 0.0016 0.0269
+ 0.0111 -0.0064 -0.0114 0.0141 -0.0239 -0.0079 0.0842 0.0459 -0.0648 -0.0352
+ 0.0119 0.0145 -0.0242 0.0128 -0.0131 0.0208 -0.0434 0.0115 -0.0482 -0.0920
+ -0.0566 0.0295 0.0205 0.0452 0.0305 0.0408 0.0249 0.0225 0.0480 0.0103
+ 0.0050 0.0329 -0.0069 0.0120 0.0461 -0.0060 -0.0195 -0.0051 -0.0226 0.0970
+ -0.0532 -0.0314 0.0361 0.0334 -0.0250 -0.0199 0.0357 0.0099 0.0315 0.0146
+ 0.0162 0.0001 0.0050 -0.0071 -0.0599 0.0304 0.0448 -0.0437 0.0666 0.0422
+ 0.0224 0.0210 0.0131 -0.0009 -0.0040 0.0236 -0.0103 -0.0693 0.0667 -0.0028
+ -0.0028 0.0234 -0.0030 -0.0098 -0.0056 -0.0588 0.0325 -0.0306 0.0157 0.0824
+ -0.0150 0.0181 -0.0314 -0.0280 0.0066 0.0350 0.0371 -0.0499 -0.0216 0.0579
+ 0.0186 -0.0115 0.0481 0.0145 -0.0390 -0.0111 -0.0245 0.0700 0.0707 0.0268
+ 0.0457 0.0144 -0.0249 0.0184 -0.0090 0.0429 0.0374 -0.0121 -0.0585 0.0050
+ -0.0144 -0.0312 -0.0086 -0.0145 -0.0394 0.0012 -0.0747 -0.0238 0.0552 -0.0041
+ 0.0197 0.0091 0.0300 0.0681 -0.0460 -0.0183 0.0031 0.0105 -0.0013 0.0106
+ -0.0106 0.0228 -0.0167 0.0270 -0.0454 0.0239 0.0329 0.0804 0.0148 -0.0122
+ -0.0013 -0.0213 -0.0091 0.0446 -0.0819 0.0332 0.0065 0.0521 0.0092 0.0246
+ -0.0272 -0.0411 -0.0116 0.0364 0.0307 -0.0209 -0.0694 -0.0261 0.0242 0.0339
+ 0.0095 0.0320 0.0391 -0.0120 -0.0429 -0.0015 0.0394 0.0458 0.0446 -0.0152
+ 0.0215 0.0017 0.0323 0.0435 0.0223 -0.0528 -0.0326 -0.0079 0.0326 -0.0337
+ 0.0113 -0.0106 0.0102 -0.0433 0.0762 0.0202 0.0773 0.0473 0.0202 -0.0192
+ -0.0142 -0.0300 -0.0323 -0.0507 -0.0188 0.0583 -0.0133 -0.0371 0.0182 -0.0453
+ -0.0194 -0.0267 0.0097 -0.0096 -0.0342 -0.0371 -0.0563 -0.0846 -0.0199 0.0363
+ -0.0071 -0.0257 -0.0614 0.0174 0.0309 0.0196 0.0026 -0.0357 -0.0750 0.0312
+ 0.0100 -0.0066 -0.0512 -0.0276 0.0047 0.0012 0.0370 -0.0546 -0.0087 -0.0600
+ -0.0183 0.0472 0.0493 0.0457 0.0039 0.0270 0.0004 -0.0277 -0.0312 -0.0484
+ 0.0001 0.0017 -0.0427 -0.0643 0.0036 -0.0225 0.0359 0.0129 -0.0554 0.0582
+ -0.0302 -0.0044 0.0017 0.0292 0.0101 -0.0688 -0.0384 -0.0162 -0.0233 -0.0349
+ 0.0210 -0.0104 0.0601 0.0254 -0.0194 0.0306 -0.0416 -0.0276 0.0364 -0.0230
+ -0.0053 -0.0353 0.0483 -0.0108 0.0658 0.0474 -0.0131 0.0064 -0.0031 -0.0390
+ 0.0014 0.0001 -0.0184 0.0683 0.0033 0.0384 -0.0511 0.0466 0.0530 0.0181
+ -0.0167 -0.0340 0.0700 0.0186 -0.0079 0.0352 -0.0031 -0.0509 -0.0289 0.0154
+ 0.0160 -0.0132 0.0177 -0.0510 0.0128 -0.0586 0.0266 -0.0333 -0.0234 -0.0743
+ -0.0225 -0.0137 -0.0289 -0.0207 0.0086 -0.0565 0.0004 0.0709 -0.0034 0.0091
+ 0.0059 -0.0095 0.0022 -0.0444 -0.0348 -0.0764 -0.0315 0.0318 0.0581 0.0199
+ 0.0221 0.0066 0.0354 -0.0416 -0.0403 0.0179 -0.0488 -0.0521 -0.0149 -0.0437
+ 0.0009 -0.0251 0.0432 0.0046 0.0217 -0.0085 -0.0086 -0.0915 0.0042 -0.0522
+ -0.0006 -0.0173 0.0162 0.0268 0.0446 0.0554 -0.0169 0.0255 -0.0259 0.0674
+ -0.0172 -0.0275 0.0095 -0.0324 -0.0134 0.0126 -0.0514 -0.0015 -0.0509 -0.0679
+ -0.0065 -0.0156 -0.0015 0.0135 0.0402 0.0955 0.0646 0.0284 0.0401 0.0262
+ -0.0070 0.0105 -0.0468 -0.0359 0.0009 -0.0027 -0.0636 -0.0538 -0.0312 0.0363
+ -0.0139 0.0451 -0.0091 0.0162 0.0621 -0.0049 0.0528 -0.0294 0.0058 -0.0150
+ -0.0055 -0.0038 -0.0025 0.0102 -0.0319 0.0348 0.0528 0.0342 0.0931 0.0517
+ 0.0367 0.0192 -0.0221 0.0310 -0.0167 -0.0004 -0.0313 -0.0630 -0.0150 0.0642
+ -0.0241 0.0472 0.0126 0.0087 -0.0460 -0.0422 0.0454 -0.0237 0.0501 0.0036
+ 0.0067 0.0350 -0.0285 -0.0260 0.0119 -0.0485 -0.0381 -0.0267 -0.0176 -0.0635
+ 0.0121 0.0203 -0.0236 -0.0380 0.0062 0.0477 0.0312 0.0154 0.0621 -0.0062
+ 0.0147 0.0012 0.0265 0.0289 -0.0247 0.0663 -0.0415 0.0201 -0.0639 -0.0118
+ -0.0047 -0.0297 0.0180 -0.0368 -0.0037 0.0025 -0.0674 0.0523 -0.0049 0.0725
+ -0.0138 -0.0289 0.0291 -0.0721 -0.0066 0.0466 0.0060 -0.0188 -0.0045 0.0039
+ -0.0206 -0.0312 -0.0213 0.0357 -0.0527 -0.0117 -0.0109 0.0070 -0.0091 -0.0623
+ 0.0159 0.0003 0.0094 0.0491 -0.0445 0.0458 0.0341 -0.0403 0.0125 -0.0376
+ -0.0030 -0.0120 0.0270 0.0196 0.0078 -0.0220 -0.0063 0.0069 -0.1155 -0.0071
+ 0.0107 0.0058 -0.0374 -0.0567 -0.0067 -0.0207 -0.0528 0.0424 0.0354 0.0544
+ -0.0006 0.0268 -0.0395 -0.0105 0.0538 0.0315 -0.0254 -0.0371 0.0691 0.0387
+ 0.0053 0.0247 0.0376 -0.0151 0.0788 0.0358 0.0177 -0.0173 0.0249 0.0386
+ -0.0155 -0.0192 -0.0087 0.0649 0.0644 -0.0048 0.0391 0.0198 0.0207 0.0028
+ 0.0063 0.0576 0.0064 0.0159 -0.0495 0.0466 -0.0360 -0.0245 0.0206 0.0045
+ -0.0292 -0.0152 -0.0133 0.0078 0.0236 0.0134 0.0321 0.0179 0.0304 0.0976
+ -0.0472 0.0330 -0.0177 0.0082 -0.0393 -0.0626 -0.0307 -0.0160 -0.0154 0.0242
+ -0.0118 -0.0289 -0.0339 -0.0022 -0.0119 -0.0292 -0.0070 -0.0407 -0.0211 0.0994
+ -0.0136 0.0393 0.0168 0.0094 -0.0201 -0.0588 0.0451 0.0313 -0.0470 -0.0180
+ -0.0376 0.0210 0.0346 0.0220 0.0013 -0.0235 -0.0347 0.0564 0.0687 0.0463
+ -0.0151 -0.0277 -0.0742 0.0173 -0.0081 0.0054 0.0073 0.0238 0.0391 -0.0050
+ 0.0288 0.0226 -0.0334 0.0183 -0.0294 -0.0766 0.0303 -0.0139 -0.0288 0.0224
+ 0.0072 0.0021 -0.0350 0.0586 0.0248 -0.0061 0.0084 -0.0499 0.0379 -0.0413
+ 0.0027 -0.0379 -0.0185 -0.0193 0.0240 0.0112 0.0531 0.0152 -0.0539 -0.0001
+ -0.0403 0.1124 0.0495 0.0017 -0.0102 0.0063 0.0255 0.0427 0.0339 0.0179
+ 0.0147 -0.0187 -0.0070 0.0201 0.0501 0.0406 0.0244 0.0229 0.0504 -0.0278
+ 0.0286 0.0061 0.0390 -0.0127 -0.0494 -0.0466 -0.0221 -0.0553 0.0394 0.0140
+ 0.0248 -0.0015 0.0216 -0.0138 -0.0370 -0.0800 0.0739 0.0365 0.0336 0.0198
+ 0.0358 0.0065 0.0026 0.0450 0.0262 -0.0277 0.0607 0.0459 -0.0322 0.0329
+ 0.0472 -0.0014 -0.0229 -0.0256 -0.0205 -0.0311 0.0209 0.0009 0.0654 0.0323
+ -0.0083 -0.0015 -0.0071 -0.0311 -0.0958 -0.0447 -0.0118 -0.0156 0.0179 0.0016
+ -0.0054 -0.0149 -0.0062 0.0064 -0.0028 0.0588 -0.0587 -0.0651 -0.0034 0.0403
+ -0.0015 -0.0163 0.0265 -0.0220 -0.0352 0.0251 0.0590 0.0119 -0.0104 0.0940
+ -0.0048 -0.0088 0.0069 0.0226 0.0261 0.0015 -0.0014 -0.0309 -0.0625 -0.1062
+ 0.0077 -0.0140 0.0336 -0.0286 0.0313 -0.0402 -0.0572 -0.0535 -0.0277 -0.0120
+ -0.0529 0.0170 -0.0417 -0.0091 0.0040 -0.0097 0.0085 0.0216 0.0661 0.0418
+ 0.0026 -0.0300 0.0471 0.0453 0.0268 -0.0077 0.0540 0.0083 -0.0211 -0.0500
+ -0.0222 -0.0034 -0.0612 -0.0342 -0.0290 -0.0053 -0.0037 0.0149 -0.0412 -0.0411
+ -0.0367 -0.0270 0.0195 -0.0235 0.0508 0.0002 0.0213 -0.0001 -0.0568 0.0487
+ 0.0032 -0.0000 -0.0413 0.0267 -0.0607 0.0555 -0.0391 -0.0442 0.0651 0.0161
+ -0.0238 0.0092 0.0057 0.0582 -0.0106 0.0010 -0.0128 -0.0633 0.0037 0.0546
+ 0.0106 0.0113 0.0629 0.0377 0.0070 -0.0085 -0.0181 0.0358 -0.0333 -0.0568
+ -0.0163 -0.0154 0.0006 0.0057 -0.0288 -0.0147 0.0418 -0.0554 -0.0491 -0.0034
+ 0.0261 0.0061 0.0060 -0.0365 0.0020 0.0751 0.0240 0.0186 -0.0222 -0.0682
+ 0.0038 0.0548 -0.0005 -0.0198 -0.0420 0.0222 0.0435 -0.0153 -0.0288 -0.0441
+ -0.0114 0.0403 -0.0115 -0.0618 -0.0497 -0.0004 -0.0247 -0.0300 -0.0054 0.0201
+ -0.0021 -0.0148 -0.0452 0.0361 0.0075 -0.0055 -0.0213 0.0548 -0.0598 0.0699
+ 0.0202 0.0480 0.0067 -0.0108 -0.0308 -0.0518 -0.0435 0.0528 0.0067 -0.0035
+ -0.0289 -0.0196 0.0430 -0.0288 0.0063 -0.0093 0.0660 0.0666 -0.0003 0.0001
+ -0.0464 -0.0121 0.0027 -0.0316 0.0434 0.0140 0.0117 -0.0558 0.0441 0.0242
+ -0.0072 -0.0009 -0.0181 -0.0302 -0.0572 -0.0376 -0.0297 -0.0474 -0.0468 -0.0785
+ -0.0017 -0.0191 0.0442 0.0479 0.0001 -0.0254 -0.0184 -0.0616 0.0675 0.0144
+ 0.0293 0.0027 -0.0168 -0.0050 0.0017 -0.0517 0.0031 0.0243 -0.0694 -0.0506
+ 0.0011 -0.0199 0.0034 -0.0641 0.0259 0.0020 0.0352 0.0282 0.0726 0.0554
+ 0.0098 -0.0183 -0.0606 -0.0063 -0.0203 -0.0588 -0.0239 -0.0324 -0.0089 -0.0059
+ 0.0253 0.0271 0.0258 0.0208 -0.0024 -0.0103 -0.0023 0.0049 0.0740 0.0714
c2->bg_est = 0.0;
c2->ex_phase = 0.0;
- for(l=1; l<MAX_AMP; l++)
+ for(l=1; l<=MAX_AMP; l++)
c2->prev_model_dec.A[l] = 0.0;
c2->prev_model_dec.Wo = TWO_PI/P_MAX;
c2->prev_model_dec.L = PI/c2->prev_model_dec.Wo;
typedef struct {
float Wo; /* fundamental frequency estimate in radians */
int L; /* number of harmonics */
- float A[MAX_AMP]; /* amplitiude of each harmonic */
- float phi[MAX_AMP]; /* phase of each harmonic */
+ float A[MAX_AMP+1]; /* amplitiude of each harmonic */
+ float phi[MAX_AMP+1]; /* phase of each harmonic */
int voiced; /* non-zero if this frame is voiced */
} MODEL;
{
int m;
float new_phi;
- COMP Ex[MAX_AMP]; /* excitation samples */
- COMP A_[MAX_AMP]; /* synthesised harmonic samples */
- COMP H[MAX_AMP]; /* LPC freq domain samples */
+ COMP Ex[MAX_AMP+1]; /* excitation samples */
+ COMP A_[MAX_AMP+1]; /* synthesised harmonic samples */
+ COMP H[MAX_AMP+1]; /* LPC freq domain samples */
float G;
float jitter = 0.0;
float r;
}
-/* Bruce Perens' funcs to load codebook files */
-
-struct codebook {
- unsigned int k;
- unsigned int log2m;
- unsigned int m;
- COMP *cb;
- unsigned int offset;
-};
-
-static const char format[] =
-"The table format must be:\n"
-"\tTwo integers describing the dimensions of the codebook.\n"
-"\tThen, enough numbers to fill the specified dimensions.\n";
-
-float get_float(FILE * in, const char * name, char * * cursor, char * buffer, int size)
-{
- for ( ; ; ) {
- char * s = *cursor;
- char c;
-
- while ( (c = *s) != '\0' && !isdigit(c) && c != '-' && c != '.' )
- s++;
-
- /* Comments start with "#" and continue to the end of the line. */
- if ( c != '\0' && c != '#' ) {
- char * end = 0;
- float f = 0;
-
- f = strtod(s, &end);
-
- if ( end != s )
- *cursor = end;
- return f;
- }
-
- if ( fgets(buffer, size, in) == NULL ) {
- fprintf(stderr, "%s: Format error. %s\n", name, format);
- exit(1);
- }
- *cursor = buffer;
- }
-}
-
-static struct codebook *load(const char * name)
-{
- FILE *file;
- char line[2048];
- char *cursor = line;
- struct codebook *b = malloc(sizeof(struct codebook));
- int i;
- int size;
- float angle;
-
- file = fopen(name, "rt");
- assert(file != NULL);
-
- *cursor = '\0';
-
- b->k = (int)get_float(file, name, &cursor, line, sizeof(line));
- b->m = (int)get_float(file, name ,&cursor, line, sizeof(line));
- size = b->k * b->m;
-
- b->cb = (COMP *)malloc(size * sizeof(COMP));
-
- for ( i = 0; i < size; i++ ) {
- angle = get_float(file, name, &cursor, line, sizeof(line));
- b->cb[i].real = cos(angle);
- b->cb[i].imag = sin(angle);
- }
-
- fclose(file);
-
- return b;
-}
-
-
-/* states for phase experiments */
-
-struct PEXP {
- float phi1;
- float phi_prev[MAX_AMP];
- float Wo_prev;
- int frames;
- float snr;
- float var;
- int var_n;
- struct codebook *vq1,*vq2,*vq3,*vq4,*vq5;
- float vq_var;
- int vq_var_n;
- MODEL prev_model;
- int state;
-};
-
-
-/*---------------------------------------------------------------------------* \
-
- phase_experiment_create()
-
- Inits states for phase quantisation experiments.
-
-\*---------------------------------------------------------------------------*/
-
-struct PEXP * phase_experiment_create() {
- struct PEXP *pexp;
- int i;
-
- pexp = (struct PEXP *)malloc(sizeof(struct PEXP));
- assert (pexp != NULL);
-
- pexp->phi1 = 0;
- for(i=0; i<MAX_AMP; i++)
- pexp->phi_prev[i] = 0.0;
- pexp->Wo_prev = 0.0;
- pexp->frames = 0;
- pexp->snr = 0.0;
- pexp->var = 0.0;
- pexp->var_n = 0;
-
- /* smoothed 10th order for 1st 1 khz */
- //pexp->vq1 = load("../unittest/ph1_10_1024.txt");
- //pexp->vq1->offset = 0;
-
- /* load experimental phase VQ */
-
- //pexp->vq1 = load("../unittest/testn1_20_1024.txt");
- pexp->vq1 = load("../unittest/test.txt");
- //pexp->vq2 = load("../unittest/testn21_40_1024.txt");
- pexp->vq2 = load("../unittest/test11_20_1024.txt");
- pexp->vq3 = load("../unittest/test21_30_1024.txt");
- pexp->vq4 = load("../unittest/test31_40_1024.txt");
- pexp->vq5 = load("../unittest/test41_60_1024.txt");
- pexp->vq1->offset = 0;
- pexp->vq2->offset = 10;
- pexp->vq3->offset = 20;
- pexp->vq4->offset = 30;
- pexp->vq5->offset = 40;
-
- pexp->vq_var = 0.0;
- pexp->vq_var_n = 0;
-
- pexp->state = 0;
-
- return pexp;
-}
-
-
-/*---------------------------------------------------------------------------* \
-
- phase_experiment_destroy()
-
-\*---------------------------------------------------------------------------*/
-
-void phase_experiment_destroy(struct PEXP *pexp) {
- assert(pexp != NULL);
- if (pexp->snr != 0.0)
- printf("snr: %4.2f dB\n", pexp->snr/pexp->frames);
- if (pexp->var != 0.0)
- printf("var...: %4.3f std dev...: %4.3f (%d non zero phases)\n",
- pexp->var/pexp->var_n, sqrt(pexp->var/pexp->var_n), pexp->var_n);
- if (pexp->vq_var != 0.0)
- printf("vq var: %4.3f vq std dev: %4.3f (%d non zero phases)\n",
- pexp->vq_var/pexp->vq_var_n, sqrt(pexp->vq_var/pexp->vq_var_n), pexp->vq_var_n);
- free(pexp);
-}
-
-
-/*---------------------------------------------------------------------------* \
-
- Various test and experimental functions ................
-
-\*---------------------------------------------------------------------------*/
-
-/* Bubblesort to find highest amplitude harmonics */
-
-struct AMPINDEX {
- float amp;
- int index;
-};
-
-static void bubbleSort(struct AMPINDEX numbers[], int array_size)
-{
- int i, j;
- struct AMPINDEX temp;
-
- for (i = (array_size - 1); i > 0; i--)
- {
- for (j = 1; j <= i; j++)
- {
- //printf("i %d j %d %f %f \n", i, j, numbers[j-1].amp, numbers[j].amp);
- if (numbers[j-1].amp < numbers[j].amp)
- {
- temp = numbers[j-1];
- numbers[j-1] = numbers[j];
- numbers[j] = temp;
- }
- }
- }
-}
-
-
-static void print_pred_error(struct PEXP *pexp, MODEL *model, int start, int end, float mag_thresh) {
- int i;
- float mag;
-
- mag = 0.0;
- for(i=start; i<=end; i++)
- mag += model->A[i]*model->A[i];
- mag = 10*log10(mag/(end-start));
-
- if (mag > mag_thresh) {
- for(i=start; i<=end; i++) {
- float pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
- float err = pred - model->phi[i];
- err = atan2(sin(err),cos(err));
- printf("%f\n",err);
- }
- //printf("\n");
- }
-
-}
-
-
-static void predict_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i;
-
- for(i=start; i<=end; i++) {
- model->phi[i] = pexp->phi_prev[i] + N*i*model->Wo;
- }
-
-}
-static float refine_Wo(struct PEXP *pexp,
- MODEL *model,
- int start,
- int end);
-
-/* Fancy state based phase prediction. Actually works OK on most utterances,
- but could use some tuning. Breaks down a bit on mmt1. */
-
-static void predict_phases_state(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i, next_state;
- float best_Wo, dWo;
-
- //best_Wo = refine_Wo(pexp, model, start, end);
- //best_Wo = (model->Wo + pexp->Wo_prev)/2.0;
- best_Wo = model->Wo;
-
- dWo = fabs(model->Wo - pexp->Wo_prev)/model->Wo;
- next_state = pexp->state;
- switch(pexp->state) {
- case 0:
- if (dWo < 0.1) {
- /* UV -> V transition, so start with phases in lock. They will
- drift a bit over voiced track which is kinda what we want, so
- we don't get clicky speech.
- */
- next_state = 1;
- for(i=start; i<=end; i++)
- pexp->phi_prev[i] = i*pexp->phi1;
- }
-
- break;
- case 1:
- if (dWo > 0.1)
- next_state = 0;
- break;
- }
- pexp->state = next_state;
-
- if (pexp->state == 0)
- for(i=start; i<=end; i++) {
- model->phi[i] = PI*(1.0 - 2.0*rand()/RAND_MAX);
- }
- else
- for(i=start; i<=end; i++) {
- model->phi[i] = pexp->phi_prev[i] + N*i*best_Wo;
- }
- printf("state %d\n", pexp->state);
-}
-
-static void struct_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i;
-
- for(i=start; i<=end; i++)
- model->phi[i] = pexp->phi1*i;
-
-}
-
-
-static void predict_phases2(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i;
- float pred, str, diff;
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*model->Wo;
- str = pexp->phi1*i;
- diff = str - pred;
- diff = atan2(sin(diff), cos(diff));
- if (diff > 0)
- pred += PI/16;
- else
- pred -= PI/16;
- model->phi[i] = pred;
- }
-
-}
-
-static void skip_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i;
-
- for(i=start; i<=end; i+=2)
- model->phi[i] = model->phi[i-1] - model->phi[i-2];
-
-}
-
-static void rand_phases(MODEL *model, int start, int end) {
- int i;
-
- for(i=start; i<=end; i++)
- model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX);
-
-}
-
-static void quant_phase(float *phase, float min, float max, int bits) {
- int levels = 1 << bits;
- int index;
- float norm, step;
-
- norm = (*phase - min)/(max - min);
- index = floor(levels*norm);
-
- //printf("phase %f norm %f index %d ", *phase, norm, index);
- if (index < 0 ) index = 0;
- if (index > (levels-1)) index = levels-1;
- //printf("index %d ", index);
- step = (max - min)/levels;
- *phase = min + step*index + 0.5*step;
- //printf("step %f phase %f\n", step, *phase);
-}
-
-static void quant_phases(MODEL *model, int start, int end, int bits) {
- int i;
-
- for(i=start; i<=end; i++) {
- quant_phase(&model->phi[i], -PI, PI, bits);
- }
-}
-
-static void fixed_bits_per_frame(struct PEXP *pexp, MODEL *model, int m, int budget) {
- int res, finished;
-
- res = 3;
- finished = 0;
-
- while(!finished) {
- if (m > model->L/2)
- res = 2;
- if (((budget - res) < 0) || (m > model->L))
- finished = 1;
- else {
- quant_phase(&model->phi[m], -PI, PI, res);
- budget -= res;
- m++;
- }
- }
- printf("m: %d L: %d budget: %d\n", m, model->L, budget);
- predict_phases(pexp, model, m, model->L);
- //rand_phases(model, m, model->L);
-}
-
-/* used to plot histogram of quantisation error, for 3 bits, 8 levels,
- should be uniform between +/- PI/8 */
-
-static void check_phase_quant(MODEL *model, float tol)
-{
- int m;
- float phi_before[MAX_AMP];
-
- for(m=1; m<=model->L; m++)
- phi_before[m] = model->phi[m];
-
- quant_phases(model, 1, model->L, 3);
-
- for(m=1; m<=model->L; m++) {
- float err = phi_before[m] - model->phi[m];
- printf("%f\n", err);
- if (fabs(err) > tol)
- exit(0);
- }
-}
-
-
-static float est_phi1(MODEL *model, int start, int end)
-{
- int m;
- float delta, s, c, phi1_est;
-
- if (end > model->L)
- end = model->L;
-
- s = c = 0.0;
- for(m=start; m<end; m++) {
- delta = model->phi[m+1] - model->phi[m];
- s += sin(delta);
- c += cos(delta);
- }
-
- phi1_est = atan2(s,c);
-
- return phi1_est;
-}
-
-static void print_phi1_pred_error(MODEL *model, int start, int end)
-{
- int m;
- float phi1_est;
-
- phi1_est = est_phi1(model, start, end);
-
- for(m=start; m<end; m++) {
- float err = model->phi[m+1] - model->phi[m] - phi1_est;
- err = atan2(sin(err),cos(err));
- printf("%f\n", err);
- }
-}
-
-
-static void first_order_band(MODEL *model, int start, int end, float phi1_est)
-{
- int m;
- float pred_err, av_pred_err;
- float c,s;
-
- s = c = 0.0;
- for(m=start; m<end; m++) {
- pred_err = model->phi[m] - phi1_est*m;
- s += sin(pred_err);
- c += cos(pred_err);
- }
-
- av_pred_err = atan2(s,c);
- for(m=start; m<end; m++) {
- model->phi[m] = av_pred_err + phi1_est*m;
- model->phi[m] = atan2(sin(model->phi[m]), cos(model->phi[m]));
- }
-
-}
-
-
-static void sub_linear(MODEL *model, int start, int end, float phi1_est)
-{
- int m;
-
- for(m=start; m<end; m++) {
- model->phi[m] = m*phi1_est;
- }
-}
-
-
-static void top_amp(struct PEXP *pexp, MODEL *model, int start, int end, int n_harm, int pred)
-{
- int removed = 0, not_removed = 0;
- int top, i, j;
- struct AMPINDEX sorted[MAX_AMP];
-
- /* sort into ascending order of amplitude */
-
- printf("\n");
- for(i=start,j=0; i<end; i++,j++) {
- sorted[j].amp = model->A[i];
- sorted[j].index = i;
- printf("%f ", model->A[i]);
- }
- bubbleSort(sorted, end-start);
-
- printf("\n");
- for(j=0; j<n_harm; j++)
- printf("%d %f\n", j, sorted[j].amp);
-
- /* keep phase of top n_harm, predict others */
-
- for(i=start; i<end; i++) {
- top = 0;
- for(j=0; j<n_harm; j++) {
- if (model->A[i] == sorted[j].amp) {
- top = 1;
- assert(i == sorted[j].index);
- }
- }
-
- #define ALTTOP
- #ifdef ALTTOP
- model->phi[i] = 0.0; /* make sure */
- if (top) {
- model->phi[i] = i*pexp->phi1;
- removed++;
- }
- else {
- model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX); // note: try rand for higher harms
- removed++;
- }
- #else
- if (!top) {
- model->phi[i] = 0.0; /* make sure */
- if (pred) {
- //model->phi[i] = pexp->phi_prev[i] + i*N*(model->Wo + pexp->Wo_prev)/2.0;
- model->phi[i] = i*model->phi[1];
- }
- else
- model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX); // note: try rand for higher harms
- removed++;
- }
- else {
- /* need to make this work thru budget of bits */
- quant_phase(&model->phi[i], -PI, PI, 3);
- not_removed++;
- }
- #endif
- }
- printf("dim: %d rem %d not_rem %d\n", end-start, removed, not_removed);
-
-}
-
-
-static void limit_prediction_error(struct PEXP *pexp, MODEL *model, int start, int end, float limit)
-{
- int i;
- float pred, pred_error, error;
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
- pred_error = pred - model->phi[i];
- pred_error -= TWO_PI*floor((pred_error+PI)/TWO_PI);
- quant_phase(&pred_error, -limit, limit, 2);
-
- error = pred - pred_error - model->phi[i];
- error -= TWO_PI*floor((error+PI)/TWO_PI);
- printf("%f\n", pred_error);
- model->phi[i] = pred - pred_error;
- }
-}
-
-
-static void quant_prediction_error(struct PEXP *pexp, MODEL *model, int start, int end, float limit)
-{
- int i;
- float pred, pred_error;
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
- pred_error = pred - model->phi[i];
- pred_error -= TWO_PI*floor((pred_error+PI)/TWO_PI);
-
- printf("%f\n", pred_error);
- model->phi[i] = pred - pred_error;
- }
-}
-
-
-static void print_sparse_pred_error(struct PEXP *pexp, MODEL *model, int start, int end, float mag_thresh)
-{
- int i, index;
- float mag, pred, error;
- float sparse_pe[MAX_AMP];
-
- mag = 0.0;
- for(i=start; i<=end; i++)
- mag += model->A[i]*model->A[i];
- mag = 10*log10(mag/(end-start));
-
- if (mag > mag_thresh) {
- for(i=0; i<MAX_AMP; i++) {
- sparse_pe[i] = 0.0;
- }
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
- error = pred - model->phi[i];
- error = atan2(sin(error),cos(error));
-
- index = MAX_AMP*i*model->Wo/PI;
- assert(index < MAX_AMP);
- sparse_pe[index] = error;
- }
-
- /* dump spare phase vector in polar format */
-
- for(i=0; i<MAX_AMP; i++)
- printf("%f ", sparse_pe[i]);
- printf("\n");
- }
-}
-
-
-static void update_snr_calc(struct PEXP *pexp, MODEL *model, float before[])
-{
- int m;
- float signal, noise, diff;
-
- signal = 0.0; noise = 0.0;
- for(m=1; m<=model->L; m++) {
- signal += model->A[m]*model->A[m];
- diff = cos(model->phi[m]) - cos(before[m]);
- noise += pow(model->A[m]*diff, 2.0);
- diff = sin(model->phi[m]) - sin(before[m]);
- noise += pow(model->A[m]*diff, 2.0);
- //printf("%f %f\n", before[m], model->phi[m]);
- }
- //printf("%f %f snr = %f\n", signal, noise, 10.0*log10(signal/noise));
- pexp->snr += 10.0*log10(signal/noise);
-}
-
-
-static void update_variance_calc(struct PEXP *pexp, MODEL *model, float before[])
-{
- int m;
- float diff;
-
- for(m=1; m<model->L; m++) {
- diff = model->phi[m] - before[m];
- diff = atan2(sin(diff), cos(diff));
- pexp->var += diff*diff;
- }
- pexp->var_n += model->L;
-}
-
-void print_vec(COMP cb[], int d, int e)
-{
- int i,j;
-
- for(j=0; j<e; j++) {
- for(i=0; i<d; i++)
- printf("%f %f ", cb[j*d+i].real, cb[j*d+i].imag);
- printf("\n");
- }
-}
-
-static COMP cconj(COMP a)
-{
- COMP res;
-
- res.real = a.real;
- res.imag = -a.imag;
-
- return res;
-}
-
-static COMP cadd(COMP a, COMP b)
-{
- COMP res;
-
- res.real = a.real + b.real;
- res.imag = a.imag + b.imag;
-
- return res;
-}
-
-static COMP cmult(COMP a, COMP b)
-{
- COMP res;
-
- res.real = a.real*b.real - a.imag*b.imag;
- res.imag = a.real*b.imag + a.imag*b.real;
-
- return res;
-}
-
-static int vq_phase(COMP cb[], COMP vec[], float weights[], int d, int e, float *se)
-{
- float error; /* current error */
- int besti; /* best index so far */
- float best_error; /* best error so far */
- int i,j;
- int ignore;
- COMP diffr;
- float diffp, metric, best_metric;
-
- besti = 0;
- best_metric = best_error = 1E32;
- for(j=0; j<e; j++) {
- error = 0.0;
- metric = 0.0;
- for(i=0; i<d; i++) {
- ignore = (vec[i].real == 0.0) && (vec[i].imag == 0.0);
- if (!ignore) {
- diffr = cmult(cb[j*d+i], cconj(vec[i]));
- diffp = atan2(diffr.imag, diffr.real);
- error += diffp*diffp;
- metric += weights[i]*weights[i]*diffp*diffp;
- //metric += weights[i]*diffp*diffp;
- //metric = log10(weights[i]*fabs(diffp));
- //printf("diffp %f metric %f\n", diffp, metric);
- //if (metric < log10(PI/(8.0*sqrt(3.0))))
- // metric = log10(PI/(8.0*sqrt(3.0)));
- }
- }
- if (metric < best_metric) {
- best_metric = metric;
- best_error = error;
- besti = j;
- }
- }
-
- *se += best_error;
-
- return(besti);
-}
-
-
-static float refine_Wo(struct PEXP *pexp,
- MODEL *model,
- int start,
- int end)
-
-{
- int i;
- float Wo_est, best_var, Wo, var, pred, error, best_Wo;
-
- /* test variance over a range of Wo values */
-
- Wo_est = (model->Wo + pexp->Wo_prev)/2.0;
- best_var = 1E32;
- for(Wo=0.97*Wo_est; Wo<=1.03*Wo_est; Wo+=0.001*Wo_est) {
-
- /* predict phase and sum differences between harmonics */
-
- var = 0.0;
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*Wo;
- error = pred - model->phi[i];
- error = atan2(sin(error),cos(error));
- var += error*error;
- }
-
- if (var < best_var) {
- best_var = var;
- best_Wo = Wo;
- }
- }
-
- return best_Wo;
-}
-
-
-static void split_vq(COMP sparse_pe_out[], struct PEXP *pexp, struct codebook *vq, float weights[], COMP sparse_pe_in[])
-{
- int i, j, non_zero, vq_ind;
-
- //printf("\n offset %d k %d m %d j: ", vq->offset, vq->k, vq->m);
- vq_ind = vq_phase(vq->cb, &sparse_pe_in[vq->offset], &weights[vq->offset], vq->k, vq->m, &pexp->vq_var);
-
- non_zero = 0;
- for(i=0, j=vq->offset; i<vq->k; i++,j++) {
- //printf("%f ", atan2(sparse_pe[i].imag, sparse_pe[i].real));
- if ((sparse_pe_in[j].real != 0.0) && (sparse_pe_in[j].imag != 0.0)) {
- //printf("%d ", j);
- sparse_pe_out[j] = vq->cb[vq->k * vq_ind + i];
- non_zero++;
- }
- }
- pexp->vq_var_n += non_zero;
-}
-
-
-static void sparse_vq_pred_error(struct PEXP *pexp,
- MODEL *model
-)
-{
- int i, index;
- float pred, error, error_q_angle, best_Wo;
- COMP sparse_pe_in[MAX_AMP], sparse_pe_out[MAX_AMP];
- float weights[MAX_AMP];
- COMP error_q_rect;
-
- best_Wo = refine_Wo(pexp, model, 1, model->L);
- //best_Wo = (model->Wo + pexp->Wo_prev)/2.0;
-
- /* transform to sparse pred error vector */
-
- for(i=0; i<MAX_AMP; i++) {
- sparse_pe_in[i].real = 0.0;
- sparse_pe_in[i].imag = 0.0;
- sparse_pe_out[i].real = 0.0;
- sparse_pe_out[i].imag = 0.0;
- }
-
- //printf("\n");
- for(i=1; i<=model->L; i++) {
- pred = pexp->phi_prev[i] + N*i*best_Wo;
- error = pred - model->phi[i];
-
- index = MAX_AMP*i*model->Wo/PI;
- assert(index < MAX_AMP);
- sparse_pe_in[index].real = cos(error);
- sparse_pe_in[index].imag = sin(error);
- sparse_pe_out[index] = sparse_pe_in[index];
- weights[index] = model->A[i];
- //printf("%d ", index);
- }
-
- /* vector quantise */
-
- split_vq(sparse_pe_out, pexp, pexp->vq1, weights, sparse_pe_in);
- split_vq(sparse_pe_out, pexp, pexp->vq2, weights, sparse_pe_in);
- split_vq(sparse_pe_out, pexp, pexp->vq3, weights, sparse_pe_in);
- split_vq(sparse_pe_out, pexp, pexp->vq4, weights, sparse_pe_in);
- split_vq(sparse_pe_out, pexp, pexp->vq5, weights, sparse_pe_in);
-
- /* transform quantised phases back */
-
- for(i=1; i<=model->L; i++) {
- pred = pexp->phi_prev[i] + N*i*best_Wo;
-
- index = MAX_AMP*i*model->Wo/PI;
- assert(index < MAX_AMP);
- error_q_rect = sparse_pe_out[index];
- error_q_angle = atan2(error_q_rect.imag, error_q_rect.real);
- model->phi[i] = pred - error_q_angle;
- model->phi[i] = atan2(sin(model->phi[i]), cos(model->phi[i]));
- }
-}
-
-
-/*
- est delta (in Wo)
- see if this gives us a better (smaller variance) prediction error
-*/
-
-static void print_pred_error_sparse_wo_correction(struct PEXP *pexp,
- MODEL *model,
- int start, int end,
- float mag_thresh)
-{
- int i, index;
- float mag, pred, error[MAX_AMP], diff, c, s, delta, err;
- float sparse_pe[MAX_AMP];
-
- mag = 0.0;
- for(i=start; i<=end; i++)
- mag += model->A[i]*model->A[i];
- mag = 10*log10(mag/(end-start));
-
- if (mag > mag_thresh) {
- for(i=0; i<MAX_AMP; i++) {
- sparse_pe[i] = 0.0;
- }
-
- /* predict phase and sum differences between harmonics */
-
- for(i=start; i<=end; i++) {
- //model->phi[i] = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0 + 0.01*N*i;
- pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
- error[i] = pred - model->phi[i];
- }
-
- /* estimate delta Wo */
-
- c = s = 0.0;
- for(i=start+1; i<=end; i++) {
- diff = error[i] - error[i-1];
- c += log(model->A[i])*cos(diff);
- s += log(model->A[i])*sin(diff);
- }
- delta = atan2(s,c)/N;
- //printf("delta %f\n",delta);
- delta = 0;
- /* now predict phases using corrected Wo */
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0 - N*i*delta;
- err = pred - model->phi[i];
- err = atan2(sin(err),cos(err));
-
- index = MAX_AMP*i*model->Wo/PI;
- assert(index < MAX_AMP);
- sparse_pe[index] = err;
- }
-
- /* dump spare phase vector in polar format */
-
- for(i=0; i<MAX_AMP; i++)
- printf("%f ", sparse_pe[i]);
- printf("\n");
-
- }
-
-}
-
-
-static void print_pred_error_sparse_wo_correction1(struct PEXP *pexp,
- MODEL *model,
- int start, int end,
- float mag_thresh)
-{
- int i, index;
- float mag, pred, best_Wo, err;
- float sparse_pe[MAX_AMP];
-
- mag = 0.0;
- for(i=start; i<=end; i++)
- mag += model->A[i]*model->A[i];
- mag = 10*log10(mag/(end-start));
-
- if (mag > mag_thresh) {
-
- best_Wo = refine_Wo(pexp, model, start, end);
-
- /* now predict phases using corrected Wo */
-
- for(i=0; i<MAX_AMP; i++) {
- sparse_pe[i] = 0.0;
- }
-
- for(i=start; i<=end; i++) {
- pred = pexp->phi_prev[i] + N*i*best_Wo;
- err = pred - model->phi[i];
- err = atan2(sin(err),cos(err));
-
- index = MAX_AMP*i*model->Wo/PI;
- assert(index < MAX_AMP);
- sparse_pe[index] = err;
- }
-
- /* dump spare phase vector in polar format */
-
- for(i=0; i<MAX_AMP; i++)
- printf("%f ", sparse_pe[i]);
- printf("\n");
-
- }
-
-}
-
-
-static void predict_phases1(struct PEXP *pexp, MODEL *model, int start, int end) {
- int i;
- float best_Wo;
-
- best_Wo = refine_Wo(pexp, model, 1, model->L);
-
- for(i=start; i<=end; i++) {
- model->phi[i] = pexp->phi_prev[i] + N*i*best_Wo;
- }
-}
-
-
-/*
- This functions tests theory that some bands can be combined together
- due to less frequency resolution at higher frequencies. This will
- reduce the amount of information we need to encode.
-*/
-
-void smooth_phase(struct PEXP *pexp, MODEL *model, int mode)
-{
- int m, i, j, index, step, v, en, nav, st;
- COMP sparse_pe_in[MAX_AMP], av;
- COMP sparse_pe_out[MAX_AMP];
- COMP smoothed[MAX_AMP];
- float best_Wo, pred, err;
- float weights[MAX_AMP];
- float avw, smoothed_weights[MAX_AMP];
- COMP smoothed_in[MAX_AMP], smoothed_out[MAX_AMP];
-
- best_Wo = refine_Wo(pexp, model, 1, model->L);
-
- for(m=0; m<MAX_AMP; m++) {
- sparse_pe_in[m].real = sparse_pe_in[m].imag = 0.0;
- sparse_pe_out[m].real = sparse_pe_out[m].imag = 0.0;
- }
-
- /* set up sparse array */
-
- for(m=1; m<=model->L; m++) {
- pred = pexp->phi_prev[m] + N*m*best_Wo;
- err = model->phi[m] - pred;
- err = atan2(sin(err),cos(err));
-
- index = MAX_AMP*m*model->Wo/PI;
- assert(index < MAX_AMP);
- sparse_pe_in[index].real = model->A[m]*cos(err);
- sparse_pe_in[index].imag = model->A[m]*sin(err);
- sparse_pe_out[index] = sparse_pe_in[index];
- weights[index] = model->A[m];
- }
-
- /* now combine samples at high frequencies to reduce dimension */
-
- step = 2;
- st = 0;
- for(i=st,v=0; i<MAX_AMP; i+=step,v++) {
-
- /* average over one band */
-
- av.real = 0.0; av.imag = 0.0; avw = 0.0; nav = 0;
- en = i+step;
- if (en > (MAX_AMP-1))
- en = MAX_AMP-1;
- for(j=i; j<en; j++) {
- if ((sparse_pe_in[j].real != 0.0) &&(sparse_pe_in[j].imag != 0.0) ) {
- av = cadd(av, sparse_pe_in[j]);
- avw += weights[j];
- nav++;
- }
- }
- if (nav) {
- smoothed[v] = av;
- smoothed_weights[v] = avw/nav;
- }
- else
- smoothed[v].real = smoothed[v].imag = 0.0;
- }
-
- if (mode == 2) {
- for(i=0; i<MAX_AMP; i++) {
- smoothed_in[i] = smoothed[i];
- smoothed_out[i] = smoothed_in[i];
- }
- split_vq(smoothed_out, pexp, pexp->vq1, smoothed_weights, smoothed_in);
- for(i=0; i<MAX_AMP; i++)
- smoothed[i] = smoothed_out[i];
- }
-
- /* set all samples to smoothed average */
-
- for(i=st,v=0; i<MAX_AMP; i+=step,v++) {
- en = i+step;
- if (en > (MAX_AMP-1))
- en = MAX_AMP-1;
- for(j=i; j<en; j++)
- sparse_pe_out[j] = smoothed[v];
- if (mode == 1)
- printf("%f ", atan2(smoothed[v].imag, smoothed[v].real));
- }
- if (mode == 1)
- printf("\n");
-
- /* convert back to Am */
-
- for(m=1; m<=model->L; m++) {
- index = MAX_AMP*m*model->Wo/PI;
- assert(index < MAX_AMP);
- pred = pexp->phi_prev[m] + N*m*best_Wo;
- err = atan2(sparse_pe_out[index].imag, sparse_pe_out[index].real);
- model->phi[m] = pred + err;
- }
-
-}
-
-/*
- Another version of a functions that tests the theory that some bands
- can be combined together due to less frequency resolution at higher
- frequencies. This will reduce the amount of information we need to
- encode.
-*/
-
-void smooth_phase2(struct PEXP *pexp, MODEL *model) {
- float m;
- float step;
- int a,b,h,i;
- float best_Wo, pred, err, s,c, phi1_;
-
- best_Wo = refine_Wo(pexp, model, 1, model->L);
-
- step = (float)model->L/30;
- printf("\nL: %d step: %3.2f am,bm: ", model->L, step);
- for(m=(float)model->L/4; m<=model->L; m+=step) {
- a = floor(m);
- b = floor(m+step);
- if (b > model->L) b = model->L;
- h = b-a;
-
- printf("%d,%d,(%d) ", a, b, h);
- c = s = 0.0;
- if (h>1) {
- for(i=a; i<b; i++) {
- pred = pexp->phi_prev[i] + N*i*best_Wo;
- err = model->phi[i] - pred;
- c += cos(err); s += sin(err);
- }
- phi1_ = atan2(s,c);
- for(i=a; i<b; i++) {
- pred = pexp->phi_prev[i] + N*i*best_Wo;
- printf("%d: %4.3f -> ", i, model->phi[i]);
- model->phi[i] = pred + phi1_;
- model->phi[i] = atan2(sin(model->phi[i]),cos(model->phi[i]));
- printf("%4.3f ", model->phi[i]);
- }
- }
- }
-}
-
-
-#define MAX_BINS 40
-//static float bins[] = {2600.0, 2800.0, 3000.0, 3200.0, 3400.0, 3600.0, 3800.0, 4000.0};
-static float bins[] = {/*
-
- 1000.0, 1100.0, 1200.0, 1300.0, 1400.0,
- 1500.0, 1600.0, 1700.0, 1800.0, 1900.0,*/
-
- 2000.0, 2400.0, 2800.0,
- 3000.0, 3400.0, 3600.0, 4000.0};
-
-void smooth_phase3(struct PEXP *pexp, MODEL *model) {
- int m, i;
- int nbins;
- int b;
- float f, best_Wo, pred, err;
- COMP av[MAX_BINS];
-
- nbins = sizeof(bins)/sizeof(float);
- best_Wo = refine_Wo(pexp, model, 1, model->L);
-
- /* clear all bins */
-
- for(i=0; i<MAX_BINS; i++) {
- av[i].real = 0.0;
- av[i].imag = 0.0;
- }
-
- /* add phases into each bin */
-
- for(m=1; m<=model->L; m++) {
- f = m*model->Wo*FS/TWO_PI;
- if (f > bins[0]) {
-
- /* find bin */
-
- for(i=0; i<nbins; i++)
- if ((f > bins[i]) && (f <= bins[i+1]))
- b = i;
- assert(b < MAX_BINS);
-
- /* est predicted phase from average */
-
- pred = pexp->phi_prev[m] + N*m*best_Wo;
- err = model->phi[m] - pred;
- av[b].real += cos(err); av[b].imag += sin(err);
- }
-
- }
-
- /* use averages to est phases */
-
- for(m=1; m<=model->L; m++) {
- f = m*model->Wo*FS/TWO_PI;
- if (f > bins[0]) {
-
- /* find bin */
-
- for(i=0; i<nbins; i++)
- if ((f > bins[i]) && (f <= bins[i+1]))
- b = i;
- assert(b < MAX_BINS);
-
- /* add predicted phase error to this bin */
-
- printf("L %d m %d f %4.f b %d\n", model->L, m, f, b);
-
- pred = pexp->phi_prev[m] + N*m*best_Wo;
- err = atan2(av[b].imag, av[b].real);
- printf(" %d: %4.3f -> ", m, model->phi[m]);
- model->phi[m] = pred + err;
- model->phi[m] = atan2(sin(model->phi[m]),cos(model->phi[m]));
- printf("%4.3f\n", model->phi[m]);
- }
- }
- printf("\n");
-}
-
-
-/*
- Try to code the phase of the largest amplitude in each band. Randomise the
- phase of the other harmonics. The theory is that only the largest harmonic
- will be audible.
-*/
-
-void cb_phase1(struct PEXP *pexp, MODEL *model) {
- int m, i;
- int nbins;
- int b;
- float f, best_Wo;
- float max_val[MAX_BINS];
- int max_ind[MAX_BINS];
-
- nbins = sizeof(bins)/sizeof(float);
- best_Wo = refine_Wo(pexp, model, 1, model->L);
-
- for(i=0; i<nbins; i++)
- max_val[i] = 0.0;
-
- /* determine max amplitude for each bin */
-
- for(m=1; m<=model->L; m++) {
- f = m*model->Wo*FS/TWO_PI;
- if (f > bins[0]) {
-
- /* find bin */
-
- for(i=0; i<nbins; i++)
- if ((f > bins[i]) && (f <= bins[i+1]))
- b = i;
- assert(b < MAX_BINS);
-
- if (model->A[m] > max_val[b]) {
- max_val[b] = model->A[m];
- max_ind[b] = m;
- }
- }
-
- }
-
- /* randomise phase of other harmonics */
-
- for(m=1; m<=model->L; m++) {
- f = m*model->Wo*FS/TWO_PI;
- if (f > bins[0]) {
-
- /* find bin */
-
- for(i=0; i<nbins; i++)
- if ((f > bins[i]) && (f <= bins[i+1]))
- b = i;
- assert(b < MAX_BINS);
-
- if (m != max_ind[b])
- model->phi[m] = pexp->phi_prev[m] + N*m*best_Wo;
- }
- }
-}
-
-
-/*
- Theory is only the phase of the envelope of signal matters within a
- Critical Band. So we estimate the position of an impulse that
- approximates the envelope of the signal.
-*/
-
-void cb_phase2(struct PEXP *pexp, MODEL *model) {
- int st, m, i, a, b, step;
- float diff,w,c,s,phi1_;
- float A[MAX_AMP];
-
- for(m=1; m<=model->L; m++) {
- A[m] = model->A[m];
- model->A[m] = 0;
- }
-
- st = 2*model->L/4;
- step = 3;
- model->phi[1] = pexp->phi_prev[1] + (pexp->Wo_prev+model->Wo)*N/2.0;
-
- printf("L=%d ", model->L);
- for(m=st; m<st+step*2; m+=step) {
- a = m; b=a+step;
- if (b > model->L)
- b = model->L;
-
- c = s = 0;
- for(i=a; i<b-1; i++) {
- printf("diff %d,%d ", i, i+1);
- diff = model->phi[i+1] - model->phi[i];
- //w = (model->A[i+1] + model->A[i])/2;
- w = 1.0;
- c += w*cos(diff); s += w*sin(diff);
- }
- phi1_ = atan2(s,c);
- printf("replacing: ");
- for(i=a; i<b; i++) {
- //model->phi[i] = i*phi1_;
- //model->phi[i] = i*model->phi[1];
- //model->phi[i] = m*(pexp->Wo_prev+model->Wo)*N/2.0;
- model->A[m] = A[m];
- printf("%d ", i);
- }
- printf(" . ");
- }
- printf("\n");
-}
-
-
-static void smooth_phase4(MODEL *model) {
- int m;
- float phi_m, phi_m_1;
-
- if (model->L > 25) {
- printf("\nL %d\n", model->L);
- for(m=model->L/2; m<=model->L; m+=2) {
- if ((m+1) <= model->L) {
- phi_m = (model->phi[m] - model->phi[m+1])/2.0;
- phi_m_1 = (model->phi[m+1] - model->phi[m])/2.0;
- model->phi[m] = phi_m;
- model->phi[m+1] = phi_m_1;
- printf("%d %4.3f %4.3f ", m, phi_m, phi_m_1);
- }
- }
- }
-
-}
-
-/* try repeating last frame, just advance phases to account for time shift */
-
-static void repeat_phases(struct PEXP *pexp, MODEL *model) {
- int m;
-
- *model = pexp->prev_model;
- for(m=1; m<=model->L; m++)
- model->phi[m] += N*m*model->Wo;
-
-}
-
-/*---------------------------------------------------------------------------*\
-
- phase_experiment()
-
- Phase quantisation experiments.
-
-\*---------------------------------------------------------------------------*/
-
-void phase_experiment(struct PEXP *pexp, MODEL *model, char *arg) {
- int m;
- float before[MAX_AMP], best_Wo;
-
- assert(pexp != NULL);
- memcpy(before, &model->phi[0], sizeof(float)*MAX_AMP);
-
- if (strcmp(arg,"q3") == 0) {
- quant_phases(model, 1, model->L, 3);
- update_snr_calc(pexp, model, before);
- update_variance_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"dec2") == 0) {
- if ((pexp->frames % 2) != 0) {
- predict_phases(pexp, model, 1, model->L);
- update_snr_calc(pexp, model, before);
- update_variance_calc(pexp, model, before);
- }
- }
-
- if (strcmp(arg,"repeat") == 0) {
- if ((pexp->frames % 2) != 0) {
- repeat_phases(pexp, model);
- update_snr_calc(pexp, model, before);
- update_variance_calc(pexp, model, before);
- }
- }
-
- if (strcmp(arg,"vq") == 0) {
- sparse_vq_pred_error(pexp, model);
- update_snr_calc(pexp, model, before);
- update_variance_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"pred") == 0)
- predict_phases_state(pexp, model, 1, model->L);
-
- if (strcmp(arg,"pred1k") == 0)
- predict_phases(pexp, model, 1, model->L/4);
-
- if (strcmp(arg,"smooth") == 0) {
- smooth_phase(pexp, model,0);
- update_snr_calc(pexp, model, before);
- }
- if (strcmp(arg,"smoothtrain") == 0)
- smooth_phase(pexp, model,1);
- if (strcmp(arg,"smoothvq") == 0) {
- smooth_phase(pexp, model,2);
- update_snr_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"smooth2") == 0)
- smooth_phase2(pexp, model);
- if (strcmp(arg,"smooth3") == 0)
- smooth_phase3(pexp, model);
- if (strcmp(arg,"smooth4") == 0)
- smooth_phase4(model);
- if (strcmp(arg,"vqsmooth3") == 0) {
- sparse_vq_pred_error(pexp, model);
- smooth_phase3(pexp, model);
- }
-
- if (strcmp(arg,"cb1") == 0) {
- cb_phase1(pexp, model);
- update_snr_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"top") == 0) {
- //top_amp(pexp, model, 1, model->L/4, 4, 1);
- //top_amp(pexp, model, model->L/4, model->L/3, 4, 1);
- //top_amp(pexp, model, model->L/3+1, model->L/2, 4, 1);
- //top_amp(pexp, model, model->L/2, model->L, 6, 1);
- //rand_phases(model, model->L/2, 3*model->L/4);
- //struct_phases(pexp, model, model->L/2, 3*model->L/4);
- //update_snr_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"pred23") == 0) {
- predict_phases2(pexp, model, model->L/2, model->L);
- update_snr_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"struct23") == 0) {
- struct_phases(pexp, model, model->L/2, 3*model->L/4 );
- update_snr_calc(pexp, model, before);
- }
-
- if (strcmp(arg,"addnoise") == 0) {
- int m;
- float max;
-
- max = 0;
- for(m=1; m<=model->L; m++)
- if (model->A[m] > max)
- max = model->A[m];
- max = 20.0*log10(max);
- for(m=1; m<=model->L; m++)
- if (20.0*log10(model->A[m]) < (max-20)) {
- model->phi[m] += (PI/4)*(1.0 -2.0*rand()/RAND_MAX);
- //printf("m %d\n", m);
- }
- }
-
- /* normalise phases */
-
- for(m=1; m<=model->L; m++)
- model->phi[m] = atan2(sin(model->phi[m]), cos(model->phi[m]));
-
- /* update states */
-
- //best_Wo = refine_Wo(pexp, model, model->L/2, model->L);
- pexp->phi1 += N*model->Wo;
-
- for(m=1; m<=model->L; m++)
- pexp->phi_prev[m] = model->phi[m];
- pexp->Wo_prev = model->Wo;
- pexp->frames++;
- pexp->prev_model = *model;
-}
-
-#ifdef OLD_STUFF
- //quant_phases(model, 1, model->L, 3);
- //update_variance_calc(pexp, model, before);
- //print_sparse_pred_error(pexp, model, 1, model->L, 40.0);
-
- //sparse_vq_pred_error(pexp, model);
-
- //quant_phases(model, model->L/4+1, model->L, 3);
-
- //predict_phases1(pexp, model, 1, model->L/4);
- //quant_phases(model, model->L/4+1, model->L, 3);
-
- //quant_phases(model, 1, model->L/8, 3);
-
- //update_snr_calc(pexp, model, before);
- //update_variance_calc(pexp, model, before);
-
- //fixed_bits_per_frame(pexp, model, 40);
- //struct_phases(pexp, model, 1, model->L/4);
- //rand_phases(model, 10, model->L);
- //for(m=1; m<=model->L; m++)
- // model->A[m] = 0.0;
- //model->A[model->L/2] = 1000;
- //repeat_phases(model, 20);
- //predict_phases(pexp, model, 1, model->L/4);
- //quant_phases(model, 1, 10, 3);
- //quant_phases(model, 10, 20, 2);
- //repeat_phases(model, 20);
- //rand_phases(model, 3*model->L/4, model->L);
- // print_phi1_pred_error(model, 1, model->L);
- //predict_phases(pexp, model, 1, model->L/4);
- //first_order_band(model, model->L/4, model->L/2);
- //first_order_band(model, model->L/2, 3*model->L/4);
- //if (fabs(model->Wo - pexp->Wo_prev)< 0.1*model->Wo)
-
- //print_pred_error(pexp, model, 1, model->L, 40.0);
- //print_sparse_pred_error(pexp, model, 1, model->L, 40.0);
-
- //phi1_est = est_phi1(model, 1, model->L/4);
- //print_phi1_pred_error(model, 1, model->L/4);
-
- //first_order_band(model, 1, model->L/4, phi1_est);
- //sub_linear(model, 1, model->L/4, phi1_est);
-
- //top_amp(pexp, model, 1, model->L/4, 4);
- //top_amp(pexp, model, model->L/4, model->L/2, 4);
-
- //first_order_band(model, 1, model->L/4, phi1_est);
- //first_order_band(model, model->L/4, model->L/2, phi1_est);
-
- //if (fabs(model->Wo - pexp->Wo_prev) > 0.2*model->Wo)
- // rand_phases(model, model->L/2, model->L);
-
- //top_amp(pexp, model, 1, model->L/4, 4);
- //top_amp(pexp, model, model->L/4, model->L/2, 8);
- //top_amp(pexp, model, model->L/4+1, model->L/2, 10, 1);
- //top_amp(pexp, model, 1, model->L/4, 10, 1);
- //top_amp(pexp, model, model->L/4+1, 3*model->L/4, 10, 1);
- //top_amp(pexp, model, 1, 3*model->L/4, 20, 1);
-
- #ifdef REAS_CAND1
- predict_phases(pexp, model, 1, model->L/4);
- top_amp(pexp, model, model->L/4+1, 3*model->L/4, 10, 1);
- rand_phases(model, 3*model->L/4+1, model->L);
- #endif
-
- #ifdef REAS_CAND2
- if ((pexp->frames % 2) == 0) {
- //printf("quant\n");
- predict_phases(pexp, model, 1, model->L/4);
- //top_amp(pexp, model, model->L/4+1, 3*model->L/4, 20, 1);
- top_amp(pexp, model, model->L/4+1, 7*model->L/8, 20, 1);
- rand_phases(model, 7*model->L/8+1, model->L);
- }
- else {
- //printf("predict\n");
- predict_phases(pexp, model, 1, model->L);
- }
- #endif
-
- //#define REAS_CAND3
- #ifdef REAS_CAND3
- if ((pexp->frames % 3) != 0) {
- printf("pred\n");
- predict_phases(pexp, model, 1, model->L);
- }
- else {
- predict_phases(pexp, model, 1, model->L/4);
- fixed_bits_per_frame(pexp, model, model->L/4+1, 60);
- }
- #endif
- //predict_phases(pexp, model, model->L/4, model->L);
-
-
- //print_pred_error(pexp, model, 1, model->L);
- //limit_prediction_error(pexp, model, model->L/2, model->L, PI/2);
-#endif
#include "kiss_fft.h"
-void phase_synth_zero_order(kiss_fft_cfg fft_dec_cfg, MODEL *model, float aks[],
- float *ex_phase, int order);
-
-struct PEXP;
-
-struct PEXP * phase_experiment_create();
-void phase_experiment_destroy(struct PEXP *pexp);
-void phase_experiment(struct PEXP *pexp, MODEL *model, char *arg);
+void phase_synth_zero_order(kiss_fft_cfg fft_dec_cfg,
+ MODEL *model,
+ float aks[],
+ float *ex_phase,
+ int order);
#endif
--- /dev/null
+/*---------------------------------------------------------------------------*\
+
+ FILE........: phaseexp.c
+ AUTHOR......: David Rowe
+ DATE CREATED: June 2012
+
+ Experimental functions for quantising, modelling and synthesising phase.
+
+\*---------------------------------------------------------------------------*/
+
+/*
+ Copyright (C) 2012 David Rowe
+
+ All rights reserved.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License version 2.1, as
+ published by the Free Software Foundation. This program is
+ distributed in the hope that it will be useful, but WITHOUT ANY
+ WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+ License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program; if not,see <http://www.gnu.org/licenses/>.
+*/
+
+#include "defines.h"
+#include "phase.h"
+#include "kiss_fft.h"
+#include "comp.h"
+#include "glottal.c"
+
+#include <assert.h>
+#include <ctype.h>
+#include <math.h>
+#include <string.h>
+#include <stdlib.h>
+
+/* Bruce Perens' funcs to load codebook files */
+
+struct codebook {
+ unsigned int k;
+ unsigned int log2m;
+ unsigned int m;
+ COMP *cb;
+ unsigned int offset;
+};
+
+static const char format[] =
+"The table format must be:\n"
+"\tTwo integers describing the dimensions of the codebook.\n"
+"\tThen, enough numbers to fill the specified dimensions.\n";
+
+float get_float(FILE * in, const char * name, char * * cursor, char * buffer, int size)
+{
+ for ( ; ; ) {
+ char * s = *cursor;
+ char c;
+
+ while ( (c = *s) != '\0' && !isdigit(c) && c != '-' && c != '.' )
+ s++;
+
+ /* Comments start with "#" and continue to the end of the line. */
+ if ( c != '\0' && c != '#' ) {
+ char * end = 0;
+ float f = 0;
+
+ f = strtod(s, &end);
+
+ if ( end != s )
+ *cursor = end;
+ return f;
+ }
+
+ if ( fgets(buffer, size, in) == NULL ) {
+ fprintf(stderr, "%s: Format error. %s\n", name, format);
+ exit(1);
+ }
+ *cursor = buffer;
+ }
+}
+
+static struct codebook *load(const char * name)
+{
+ FILE *file;
+ char line[2048];
+ char *cursor = line;
+ struct codebook *b = malloc(sizeof(struct codebook));
+ int i;
+ int size;
+ float angle;
+
+ file = fopen(name, "rt");
+ assert(file != NULL);
+
+ *cursor = '\0';
+
+ b->k = (int)get_float(file, name, &cursor, line, sizeof(line));
+ b->m = (int)get_float(file, name ,&cursor, line, sizeof(line));
+ size = b->k * b->m;
+
+ b->cb = (COMP *)malloc(size * sizeof(COMP));
+
+ for ( i = 0; i < size; i++ ) {
+ angle = get_float(file, name, &cursor, line, sizeof(line));
+ b->cb[i].real = cos(angle);
+ b->cb[i].imag = sin(angle);
+ }
+
+ fclose(file);
+
+ return b;
+}
+
+
+/* states for phase experiments */
+
+struct PEXP {
+ float phi1;
+ float phi_prev[MAX_AMP];
+ float Wo_prev;
+ int frames;
+ float snr;
+ float var;
+ int var_n;
+ struct codebook *vq1,*vq2,*vq3,*vq4,*vq5;
+ float vq_var;
+ int vq_var_n;
+ MODEL prev_model;
+ int state;
+};
+
+
+/*---------------------------------------------------------------------------* \
+
+ phase_experiment_create()
+
+ Inits states for phase quantisation experiments.
+
+\*---------------------------------------------------------------------------*/
+
+struct PEXP * phase_experiment_create() {
+ struct PEXP *pexp;
+ int i;
+
+ pexp = (struct PEXP *)malloc(sizeof(struct PEXP));
+ assert (pexp != NULL);
+
+ pexp->phi1 = 0;
+ for(i=0; i<MAX_AMP; i++)
+ pexp->phi_prev[i] = 0.0;
+ pexp->Wo_prev = 0.0;
+ pexp->frames = 0;
+ pexp->snr = 0.0;
+ pexp->var = 0.0;
+ pexp->var_n = 0;
+
+ /* smoothed 10th order for 1st 1 khz */
+ //pexp->vq1 = load("../unittest/ph1_10_1024.txt");
+ //pexp->vq1->offset = 0;
+
+ /* load experimental phase VQ */
+
+ //pexp->vq1 = load("../unittest/testn1_20_1024.txt");
+ pexp->vq1 = load("../unittest/test.txt");
+ //pexp->vq2 = load("../unittest/testn21_40_1024.txt");
+ pexp->vq2 = load("../unittest/test11_20_1024.txt");
+ pexp->vq3 = load("../unittest/test21_30_1024.txt");
+ pexp->vq4 = load("../unittest/test31_40_1024.txt");
+ pexp->vq5 = load("../unittest/test41_60_1024.txt");
+ pexp->vq1->offset = 0;
+ pexp->vq2->offset = 10;
+ pexp->vq3->offset = 20;
+ pexp->vq4->offset = 30;
+ pexp->vq5->offset = 40;
+
+ pexp->vq_var = 0.0;
+ pexp->vq_var_n = 0;
+
+ pexp->state = 0;
+
+ return pexp;
+}
+
+
+/*---------------------------------------------------------------------------* \
+
+ phase_experiment_destroy()
+
+\*---------------------------------------------------------------------------*/
+
+void phase_experiment_destroy(struct PEXP *pexp) {
+ assert(pexp != NULL);
+ if (pexp->snr != 0.0)
+ printf("snr: %4.2f dB\n", pexp->snr/pexp->frames);
+ if (pexp->var != 0.0)
+ printf("var...: %4.3f std dev...: %4.3f (%d non zero phases)\n",
+ pexp->var/pexp->var_n, sqrt(pexp->var/pexp->var_n), pexp->var_n);
+ if (pexp->vq_var != 0.0)
+ printf("vq var: %4.3f vq std dev: %4.3f (%d non zero phases)\n",
+ pexp->vq_var/pexp->vq_var_n, sqrt(pexp->vq_var/pexp->vq_var_n), pexp->vq_var_n);
+ free(pexp);
+}
+
+
+/*---------------------------------------------------------------------------* \
+
+ Various test and experimental functions ................
+
+\*---------------------------------------------------------------------------*/
+
+/* Bubblesort to find highest amplitude harmonics */
+
+struct AMPINDEX {
+ float amp;
+ int index;
+};
+
+static void bubbleSort(struct AMPINDEX numbers[], int array_size)
+{
+ int i, j;
+ struct AMPINDEX temp;
+
+ for (i = (array_size - 1); i > 0; i--)
+ {
+ for (j = 1; j <= i; j++)
+ {
+ //printf("i %d j %d %f %f \n", i, j, numbers[j-1].amp, numbers[j].amp);
+ if (numbers[j-1].amp < numbers[j].amp)
+ {
+ temp = numbers[j-1];
+ numbers[j-1] = numbers[j];
+ numbers[j] = temp;
+ }
+ }
+ }
+}
+
+
+static void print_pred_error(struct PEXP *pexp, MODEL *model, int start, int end, float mag_thresh) {
+ int i;
+ float mag;
+
+ mag = 0.0;
+ for(i=start; i<=end; i++)
+ mag += model->A[i]*model->A[i];
+ mag = 10*log10(mag/(end-start));
+
+ if (mag > mag_thresh) {
+ for(i=start; i<=end; i++) {
+ float pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
+ float err = pred - model->phi[i];
+ err = atan2(sin(err),cos(err));
+ printf("%f\n",err);
+ }
+ //printf("\n");
+ }
+
+}
+
+
+static void predict_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i;
+
+ for(i=start; i<=end; i++) {
+ model->phi[i] = pexp->phi_prev[i] + N*i*model->Wo;
+ }
+
+}
+static float refine_Wo(struct PEXP *pexp,
+ MODEL *model,
+ int start,
+ int end);
+
+/* Fancy state based phase prediction. Actually works OK on most utterances,
+ but could use some tuning. Breaks down a bit on mmt1. */
+
+static void predict_phases_state(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i, next_state;
+ float best_Wo, dWo;
+
+ //best_Wo = refine_Wo(pexp, model, start, end);
+ //best_Wo = (model->Wo + pexp->Wo_prev)/2.0;
+ best_Wo = model->Wo;
+
+ dWo = fabs(model->Wo - pexp->Wo_prev)/model->Wo;
+ next_state = pexp->state;
+ switch(pexp->state) {
+ case 0:
+ if (dWo < 0.1) {
+ /* UV -> V transition, so start with phases in lock. They will
+ drift a bit over voiced track which is kinda what we want, so
+ we don't get clicky speech.
+ */
+ next_state = 1;
+ for(i=start; i<=end; i++)
+ pexp->phi_prev[i] = i*pexp->phi1;
+ }
+
+ break;
+ case 1:
+ if (dWo > 0.1)
+ next_state = 0;
+ break;
+ }
+ pexp->state = next_state;
+
+ if (pexp->state == 0)
+ for(i=start; i<=end; i++) {
+ model->phi[i] = PI*(1.0 - 2.0*rand()/RAND_MAX);
+ }
+ else
+ for(i=start; i<=end; i++) {
+ model->phi[i] = pexp->phi_prev[i] + N*i*best_Wo;
+ }
+ printf("state %d\n", pexp->state);
+}
+
+static void struct_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i;
+
+ for(i=start; i<=end; i++)
+ model->phi[i] = pexp->phi1*i;
+
+}
+
+
+static void predict_phases2(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i;
+ float pred, str, diff;
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*model->Wo;
+ str = pexp->phi1*i;
+ diff = str - pred;
+ diff = atan2(sin(diff), cos(diff));
+ if (diff > 0)
+ pred += PI/16;
+ else
+ pred -= PI/16;
+ model->phi[i] = pred;
+ }
+
+}
+
+static void skip_phases(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i;
+
+ for(i=start; i<=end; i+=2)
+ model->phi[i] = model->phi[i-1] - model->phi[i-2];
+
+}
+
+static void rand_phases(MODEL *model, int start, int end) {
+ int i;
+
+ for(i=start; i<=end; i++)
+ model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX);
+
+}
+
+static void quant_phase(float *phase, float min, float max, int bits) {
+ int levels = 1 << bits;
+ int index;
+ float norm, step;
+
+ norm = (*phase - min)/(max - min);
+ index = floor(levels*norm);
+
+ //printf("phase %f norm %f index %d ", *phase, norm, index);
+ if (index < 0 ) index = 0;
+ if (index > (levels-1)) index = levels-1;
+ //printf("index %d ", index);
+ step = (max - min)/levels;
+ *phase = min + step*index + 0.5*step;
+ //printf("step %f phase %f\n", step, *phase);
+}
+
+static void quant_phases(MODEL *model, int start, int end, int bits) {
+ int i;
+
+ for(i=start; i<=end; i++) {
+ quant_phase(&model->phi[i], -PI, PI, bits);
+ }
+}
+
+static void fixed_bits_per_frame(struct PEXP *pexp, MODEL *model, int m, int budget) {
+ int res, finished;
+
+ res = 3;
+ finished = 0;
+
+ while(!finished) {
+ if (m > model->L/2)
+ res = 2;
+ if (((budget - res) < 0) || (m > model->L))
+ finished = 1;
+ else {
+ quant_phase(&model->phi[m], -PI, PI, res);
+ budget -= res;
+ m++;
+ }
+ }
+ printf("m: %d L: %d budget: %d\n", m, model->L, budget);
+ predict_phases(pexp, model, m, model->L);
+ //rand_phases(model, m, model->L);
+}
+
+/* used to plot histogram of quantisation error, for 3 bits, 8 levels,
+ should be uniform between +/- PI/8 */
+
+static void check_phase_quant(MODEL *model, float tol)
+{
+ int m;
+ float phi_before[MAX_AMP];
+
+ for(m=1; m<=model->L; m++)
+ phi_before[m] = model->phi[m];
+
+ quant_phases(model, 1, model->L, 3);
+
+ for(m=1; m<=model->L; m++) {
+ float err = phi_before[m] - model->phi[m];
+ printf("%f\n", err);
+ if (fabs(err) > tol)
+ exit(0);
+ }
+}
+
+
+static float est_phi1(MODEL *model, int start, int end)
+{
+ int m;
+ float delta, s, c, phi1_est;
+
+ if (end > model->L)
+ end = model->L;
+
+ s = c = 0.0;
+ for(m=start; m<end; m++) {
+ delta = model->phi[m+1] - model->phi[m];
+ s += sin(delta);
+ c += cos(delta);
+ }
+
+ phi1_est = atan2(s,c);
+
+ return phi1_est;
+}
+
+static void print_phi1_pred_error(MODEL *model, int start, int end)
+{
+ int m;
+ float phi1_est;
+
+ phi1_est = est_phi1(model, start, end);
+
+ for(m=start; m<end; m++) {
+ float err = model->phi[m+1] - model->phi[m] - phi1_est;
+ err = atan2(sin(err),cos(err));
+ printf("%f\n", err);
+ }
+}
+
+
+static void first_order_band(MODEL *model, int start, int end, float phi1_est)
+{
+ int m;
+ float pred_err, av_pred_err;
+ float c,s;
+
+ s = c = 0.0;
+ for(m=start; m<end; m++) {
+ pred_err = model->phi[m] - phi1_est*m;
+ s += sin(pred_err);
+ c += cos(pred_err);
+ }
+
+ av_pred_err = atan2(s,c);
+ for(m=start; m<end; m++) {
+ model->phi[m] = av_pred_err + phi1_est*m;
+ model->phi[m] = atan2(sin(model->phi[m]), cos(model->phi[m]));
+ }
+
+}
+
+
+static void sub_linear(MODEL *model, int start, int end, float phi1_est)
+{
+ int m;
+
+ for(m=start; m<end; m++) {
+ model->phi[m] = m*phi1_est;
+ }
+}
+
+
+static void top_amp(struct PEXP *pexp, MODEL *model, int start, int end, int n_harm, int pred)
+{
+ int removed = 0, not_removed = 0;
+ int top, i, j;
+ struct AMPINDEX sorted[MAX_AMP];
+
+ /* sort into ascending order of amplitude */
+
+ printf("\n");
+ for(i=start,j=0; i<end; i++,j++) {
+ sorted[j].amp = model->A[i];
+ sorted[j].index = i;
+ printf("%f ", model->A[i]);
+ }
+ bubbleSort(sorted, end-start);
+
+ printf("\n");
+ for(j=0; j<n_harm; j++)
+ printf("%d %f\n", j, sorted[j].amp);
+
+ /* keep phase of top n_harm, predict others */
+
+ for(i=start; i<end; i++) {
+ top = 0;
+ for(j=0; j<n_harm; j++) {
+ if (model->A[i] == sorted[j].amp) {
+ top = 1;
+ assert(i == sorted[j].index);
+ }
+ }
+
+ #define ALTTOP
+ #ifdef ALTTOP
+ model->phi[i] = 0.0; /* make sure */
+ if (top) {
+ model->phi[i] = i*pexp->phi1;
+ removed++;
+ }
+ else {
+ model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX); // note: try rand for higher harms
+ removed++;
+ }
+ #else
+ if (!top) {
+ model->phi[i] = 0.0; /* make sure */
+ if (pred) {
+ //model->phi[i] = pexp->phi_prev[i] + i*N*(model->Wo + pexp->Wo_prev)/2.0;
+ model->phi[i] = i*model->phi[1];
+ }
+ else
+ model->phi[i] = PI*(1.0 - 2.0*(float)rand()/RAND_MAX); // note: try rand for higher harms
+ removed++;
+ }
+ else {
+ /* need to make this work thru budget of bits */
+ quant_phase(&model->phi[i], -PI, PI, 3);
+ not_removed++;
+ }
+ #endif
+ }
+ printf("dim: %d rem %d not_rem %d\n", end-start, removed, not_removed);
+
+}
+
+
+static void limit_prediction_error(struct PEXP *pexp, MODEL *model, int start, int end, float limit)
+{
+ int i;
+ float pred, pred_error, error;
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
+ pred_error = pred - model->phi[i];
+ pred_error -= TWO_PI*floor((pred_error+PI)/TWO_PI);
+ quant_phase(&pred_error, -limit, limit, 2);
+
+ error = pred - pred_error - model->phi[i];
+ error -= TWO_PI*floor((error+PI)/TWO_PI);
+ printf("%f\n", pred_error);
+ model->phi[i] = pred - pred_error;
+ }
+}
+
+
+static void quant_prediction_error(struct PEXP *pexp, MODEL *model, int start, int end, float limit)
+{
+ int i;
+ float pred, pred_error;
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
+ pred_error = pred - model->phi[i];
+ pred_error -= TWO_PI*floor((pred_error+PI)/TWO_PI);
+
+ printf("%f\n", pred_error);
+ model->phi[i] = pred - pred_error;
+ }
+}
+
+
+static void print_sparse_pred_error(struct PEXP *pexp, MODEL *model, int start, int end, float mag_thresh)
+{
+ int i, index;
+ float mag, pred, error;
+ float sparse_pe[MAX_AMP];
+
+ mag = 0.0;
+ for(i=start; i<=end; i++)
+ mag += model->A[i]*model->A[i];
+ mag = 10*log10(mag/(end-start));
+
+ if (mag > mag_thresh) {
+ for(i=0; i<MAX_AMP; i++) {
+ sparse_pe[i] = 0.0;
+ }
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
+ error = pred - model->phi[i];
+ error = atan2(sin(error),cos(error));
+
+ index = MAX_AMP*i*model->Wo/PI;
+ assert(index < MAX_AMP);
+ sparse_pe[index] = error;
+ }
+
+ /* dump spare phase vector in polar format */
+
+ for(i=0; i<MAX_AMP; i++)
+ printf("%f ", sparse_pe[i]);
+ printf("\n");
+ }
+}
+
+
+static void update_snr_calc(struct PEXP *pexp, MODEL *model, float before[])
+{
+ int m;
+ float signal, noise, diff;
+
+ signal = 0.0; noise = 0.0;
+ for(m=1; m<=model->L; m++) {
+ signal += model->A[m]*model->A[m];
+ diff = cos(model->phi[m]) - cos(before[m]);
+ noise += pow(model->A[m]*diff, 2.0);
+ diff = sin(model->phi[m]) - sin(before[m]);
+ noise += pow(model->A[m]*diff, 2.0);
+ //printf("%f %f\n", before[m], model->phi[m]);
+ }
+ //printf("%f %f snr = %f\n", signal, noise, 10.0*log10(signal/noise));
+ pexp->snr += 10.0*log10(signal/noise);
+}
+
+
+static void update_variance_calc(struct PEXP *pexp, MODEL *model, float before[])
+{
+ int m;
+ float diff;
+
+ for(m=1; m<model->L; m++) {
+ diff = model->phi[m] - before[m];
+ diff = atan2(sin(diff), cos(diff));
+ pexp->var += diff*diff;
+ }
+ pexp->var_n += model->L;
+}
+
+void print_vec(COMP cb[], int d, int e)
+{
+ int i,j;
+
+ for(j=0; j<e; j++) {
+ for(i=0; i<d; i++)
+ printf("%f %f ", cb[j*d+i].real, cb[j*d+i].imag);
+ printf("\n");
+ }
+}
+
+static COMP cconj(COMP a)
+{
+ COMP res;
+
+ res.real = a.real;
+ res.imag = -a.imag;
+
+ return res;
+}
+
+static COMP cadd(COMP a, COMP b)
+{
+ COMP res;
+
+ res.real = a.real + b.real;
+ res.imag = a.imag + b.imag;
+
+ return res;
+}
+
+static COMP cmult(COMP a, COMP b)
+{
+ COMP res;
+
+ res.real = a.real*b.real - a.imag*b.imag;
+ res.imag = a.real*b.imag + a.imag*b.real;
+
+ return res;
+}
+
+static int vq_phase(COMP cb[], COMP vec[], float weights[], int d, int e, float *se)
+{
+ float error; /* current error */
+ int besti; /* best index so far */
+ float best_error; /* best error so far */
+ int i,j;
+ int ignore;
+ COMP diffr;
+ float diffp, metric, best_metric;
+
+ besti = 0;
+ best_metric = best_error = 1E32;
+ for(j=0; j<e; j++) {
+ error = 0.0;
+ metric = 0.0;
+ for(i=0; i<d; i++) {
+ ignore = (vec[i].real == 0.0) && (vec[i].imag == 0.0);
+ if (!ignore) {
+ diffr = cmult(cb[j*d+i], cconj(vec[i]));
+ diffp = atan2(diffr.imag, diffr.real);
+ error += diffp*diffp;
+ metric += weights[i]*weights[i]*diffp*diffp;
+ //metric += weights[i]*diffp*diffp;
+ //metric = log10(weights[i]*fabs(diffp));
+ //printf("diffp %f metric %f\n", diffp, metric);
+ //if (metric < log10(PI/(8.0*sqrt(3.0))))
+ // metric = log10(PI/(8.0*sqrt(3.0)));
+ }
+ }
+ if (metric < best_metric) {
+ best_metric = metric;
+ best_error = error;
+ besti = j;
+ }
+ }
+
+ *se += best_error;
+
+ return(besti);
+}
+
+
+static float refine_Wo(struct PEXP *pexp,
+ MODEL *model,
+ int start,
+ int end)
+
+{
+ int i;
+ float Wo_est, best_var, Wo, var, pred, error, best_Wo;
+
+ /* test variance over a range of Wo values */
+
+ Wo_est = (model->Wo + pexp->Wo_prev)/2.0;
+ best_var = 1E32;
+ for(Wo=0.97*Wo_est; Wo<=1.03*Wo_est; Wo+=0.001*Wo_est) {
+
+ /* predict phase and sum differences between harmonics */
+
+ var = 0.0;
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*Wo;
+ error = pred - model->phi[i];
+ error = atan2(sin(error),cos(error));
+ var += error*error;
+ }
+
+ if (var < best_var) {
+ best_var = var;
+ best_Wo = Wo;
+ }
+ }
+
+ return best_Wo;
+}
+
+
+static void split_vq(COMP sparse_pe_out[], struct PEXP *pexp, struct codebook *vq, float weights[], COMP sparse_pe_in[])
+{
+ int i, j, non_zero, vq_ind;
+
+ //printf("\n offset %d k %d m %d j: ", vq->offset, vq->k, vq->m);
+ vq_ind = vq_phase(vq->cb, &sparse_pe_in[vq->offset], &weights[vq->offset], vq->k, vq->m, &pexp->vq_var);
+
+ non_zero = 0;
+ for(i=0, j=vq->offset; i<vq->k; i++,j++) {
+ //printf("%f ", atan2(sparse_pe[i].imag, sparse_pe[i].real));
+ if ((sparse_pe_in[j].real != 0.0) && (sparse_pe_in[j].imag != 0.0)) {
+ //printf("%d ", j);
+ sparse_pe_out[j] = vq->cb[vq->k * vq_ind + i];
+ non_zero++;
+ }
+ }
+ pexp->vq_var_n += non_zero;
+}
+
+
+static void sparse_vq_pred_error(struct PEXP *pexp,
+ MODEL *model
+)
+{
+ int i, index;
+ float pred, error, error_q_angle, best_Wo;
+ COMP sparse_pe_in[MAX_AMP], sparse_pe_out[MAX_AMP];
+ float weights[MAX_AMP];
+ COMP error_q_rect;
+
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+ //best_Wo = (model->Wo + pexp->Wo_prev)/2.0;
+
+ /* transform to sparse pred error vector */
+
+ for(i=0; i<MAX_AMP; i++) {
+ sparse_pe_in[i].real = 0.0;
+ sparse_pe_in[i].imag = 0.0;
+ sparse_pe_out[i].real = 0.0;
+ sparse_pe_out[i].imag = 0.0;
+ }
+
+ //printf("\n");
+ for(i=1; i<=model->L; i++) {
+ pred = pexp->phi_prev[i] + N*i*best_Wo;
+ error = pred - model->phi[i];
+
+ index = MAX_AMP*i*model->Wo/PI;
+ assert(index < MAX_AMP);
+ sparse_pe_in[index].real = cos(error);
+ sparse_pe_in[index].imag = sin(error);
+ sparse_pe_out[index] = sparse_pe_in[index];
+ weights[index] = model->A[i];
+ //printf("%d ", index);
+ }
+
+ /* vector quantise */
+
+ split_vq(sparse_pe_out, pexp, pexp->vq1, weights, sparse_pe_in);
+ split_vq(sparse_pe_out, pexp, pexp->vq2, weights, sparse_pe_in);
+ split_vq(sparse_pe_out, pexp, pexp->vq3, weights, sparse_pe_in);
+ split_vq(sparse_pe_out, pexp, pexp->vq4, weights, sparse_pe_in);
+ split_vq(sparse_pe_out, pexp, pexp->vq5, weights, sparse_pe_in);
+
+ /* transform quantised phases back */
+
+ for(i=1; i<=model->L; i++) {
+ pred = pexp->phi_prev[i] + N*i*best_Wo;
+
+ index = MAX_AMP*i*model->Wo/PI;
+ assert(index < MAX_AMP);
+ error_q_rect = sparse_pe_out[index];
+ error_q_angle = atan2(error_q_rect.imag, error_q_rect.real);
+ model->phi[i] = pred - error_q_angle;
+ model->phi[i] = atan2(sin(model->phi[i]), cos(model->phi[i]));
+ }
+}
+
+
+/*
+ est delta (in Wo)
+ see if this gives us a better (smaller variance) prediction error
+*/
+
+static void print_pred_error_sparse_wo_correction(struct PEXP *pexp,
+ MODEL *model,
+ int start, int end,
+ float mag_thresh)
+{
+ int i, index;
+ float mag, pred, error[MAX_AMP], diff, c, s, delta, err;
+ float sparse_pe[MAX_AMP];
+
+ mag = 0.0;
+ for(i=start; i<=end; i++)
+ mag += model->A[i]*model->A[i];
+ mag = 10*log10(mag/(end-start));
+
+ if (mag > mag_thresh) {
+ for(i=0; i<MAX_AMP; i++) {
+ sparse_pe[i] = 0.0;
+ }
+
+ /* predict phase and sum differences between harmonics */
+
+ for(i=start; i<=end; i++) {
+ //model->phi[i] = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0 + 0.01*N*i;
+ pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0;
+ error[i] = pred - model->phi[i];
+ }
+
+ /* estimate delta Wo */
+
+ c = s = 0.0;
+ for(i=start+1; i<=end; i++) {
+ diff = error[i] - error[i-1];
+ c += log(model->A[i])*cos(diff);
+ s += log(model->A[i])*sin(diff);
+ }
+ delta = atan2(s,c)/N;
+ //printf("delta %f\n",delta);
+ delta = 0;
+ /* now predict phases using corrected Wo */
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*(model->Wo + pexp->Wo_prev)/2.0 - N*i*delta;
+ err = pred - model->phi[i];
+ err = atan2(sin(err),cos(err));
+
+ index = MAX_AMP*i*model->Wo/PI;
+ assert(index < MAX_AMP);
+ sparse_pe[index] = err;
+ }
+
+ /* dump spare phase vector in polar format */
+
+ for(i=0; i<MAX_AMP; i++)
+ printf("%f ", sparse_pe[i]);
+ printf("\n");
+
+ }
+
+}
+
+
+static void print_pred_error_sparse_wo_correction1(struct PEXP *pexp,
+ MODEL *model,
+ int start, int end,
+ float mag_thresh)
+{
+ int i, index;
+ float mag, pred, best_Wo, err;
+ float sparse_pe[MAX_AMP];
+
+ mag = 0.0;
+ for(i=start; i<=end; i++)
+ mag += model->A[i]*model->A[i];
+ mag = 10*log10(mag/(end-start));
+
+ if (mag > mag_thresh) {
+
+ best_Wo = refine_Wo(pexp, model, start, end);
+
+ /* now predict phases using corrected Wo */
+
+ for(i=0; i<MAX_AMP; i++) {
+ sparse_pe[i] = 0.0;
+ }
+
+ for(i=start; i<=end; i++) {
+ pred = pexp->phi_prev[i] + N*i*best_Wo;
+ err = pred - model->phi[i];
+ err = atan2(sin(err),cos(err));
+
+ index = MAX_AMP*i*model->Wo/PI;
+ assert(index < MAX_AMP);
+ sparse_pe[index] = err;
+ }
+
+ /* dump spare phase vector in polar format */
+
+ for(i=0; i<MAX_AMP; i++)
+ printf("%f ", sparse_pe[i]);
+ printf("\n");
+
+ }
+
+}
+
+
+static void predict_phases1(struct PEXP *pexp, MODEL *model, int start, int end) {
+ int i;
+ float best_Wo;
+
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+
+ for(i=start; i<=end; i++) {
+ model->phi[i] = pexp->phi_prev[i] + N*i*best_Wo;
+ }
+}
+
+
+/*
+ This functions tests theory that some bands can be combined together
+ due to less frequency resolution at higher frequencies. This will
+ reduce the amount of information we need to encode.
+*/
+
+void smooth_phase(struct PEXP *pexp, MODEL *model, int mode)
+{
+ int m, i, j, index, step, v, en, nav, st;
+ COMP sparse_pe_in[MAX_AMP], av;
+ COMP sparse_pe_out[MAX_AMP];
+ COMP smoothed[MAX_AMP];
+ float best_Wo, pred, err;
+ float weights[MAX_AMP];
+ float avw, smoothed_weights[MAX_AMP];
+ COMP smoothed_in[MAX_AMP], smoothed_out[MAX_AMP];
+
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+
+ for(m=0; m<MAX_AMP; m++) {
+ sparse_pe_in[m].real = sparse_pe_in[m].imag = 0.0;
+ sparse_pe_out[m].real = sparse_pe_out[m].imag = 0.0;
+ }
+
+ /* set up sparse array */
+
+ for(m=1; m<=model->L; m++) {
+ pred = pexp->phi_prev[m] + N*m*best_Wo;
+ err = model->phi[m] - pred;
+ err = atan2(sin(err),cos(err));
+
+ index = MAX_AMP*m*model->Wo/PI;
+ assert(index < MAX_AMP);
+ sparse_pe_in[index].real = model->A[m]*cos(err);
+ sparse_pe_in[index].imag = model->A[m]*sin(err);
+ sparse_pe_out[index] = sparse_pe_in[index];
+ weights[index] = model->A[m];
+ }
+
+ /* now combine samples at high frequencies to reduce dimension */
+
+ step = 2;
+ st = 0;
+ for(i=st,v=0; i<MAX_AMP; i+=step,v++) {
+
+ /* average over one band */
+
+ av.real = 0.0; av.imag = 0.0; avw = 0.0; nav = 0;
+ en = i+step;
+ if (en > (MAX_AMP-1))
+ en = MAX_AMP-1;
+ for(j=i; j<en; j++) {
+ if ((sparse_pe_in[j].real != 0.0) &&(sparse_pe_in[j].imag != 0.0) ) {
+ av = cadd(av, sparse_pe_in[j]);
+ avw += weights[j];
+ nav++;
+ }
+ }
+ if (nav) {
+ smoothed[v] = av;
+ smoothed_weights[v] = avw/nav;
+ }
+ else
+ smoothed[v].real = smoothed[v].imag = 0.0;
+ }
+
+ if (mode == 2) {
+ for(i=0; i<MAX_AMP; i++) {
+ smoothed_in[i] = smoothed[i];
+ smoothed_out[i] = smoothed_in[i];
+ }
+ split_vq(smoothed_out, pexp, pexp->vq1, smoothed_weights, smoothed_in);
+ for(i=0; i<MAX_AMP; i++)
+ smoothed[i] = smoothed_out[i];
+ }
+
+ /* set all samples to smoothed average */
+
+ for(i=st,v=0; i<MAX_AMP; i+=step,v++) {
+ en = i+step;
+ if (en > (MAX_AMP-1))
+ en = MAX_AMP-1;
+ for(j=i; j<en; j++)
+ sparse_pe_out[j] = smoothed[v];
+ if (mode == 1)
+ printf("%f ", atan2(smoothed[v].imag, smoothed[v].real));
+ }
+ if (mode == 1)
+ printf("\n");
+
+ /* convert back to Am */
+
+ for(m=1; m<=model->L; m++) {
+ index = MAX_AMP*m*model->Wo/PI;
+ assert(index < MAX_AMP);
+ pred = pexp->phi_prev[m] + N*m*best_Wo;
+ err = atan2(sparse_pe_out[index].imag, sparse_pe_out[index].real);
+ model->phi[m] = pred + err;
+ }
+
+}
+
+/*
+ Another version of a functions that tests the theory that some bands
+ can be combined together due to less frequency resolution at higher
+ frequencies. This will reduce the amount of information we need to
+ encode.
+*/
+
+void smooth_phase2(struct PEXP *pexp, MODEL *model) {
+ float m;
+ float step;
+ int a,b,h,i;
+ float best_Wo, pred, err, s,c, phi1_;
+
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+
+ step = (float)model->L/30;
+ printf("\nL: %d step: %3.2f am,bm: ", model->L, step);
+ for(m=(float)model->L/4; m<=model->L; m+=step) {
+ a = floor(m);
+ b = floor(m+step);
+ if (b > model->L) b = model->L;
+ h = b-a;
+
+ printf("%d,%d,(%d) ", a, b, h);
+ c = s = 0.0;
+ if (h>1) {
+ for(i=a; i<b; i++) {
+ pred = pexp->phi_prev[i] + N*i*best_Wo;
+ err = model->phi[i] - pred;
+ c += cos(err); s += sin(err);
+ }
+ phi1_ = atan2(s,c);
+ for(i=a; i<b; i++) {
+ pred = pexp->phi_prev[i] + N*i*best_Wo;
+ printf("%d: %4.3f -> ", i, model->phi[i]);
+ model->phi[i] = pred + phi1_;
+ model->phi[i] = atan2(sin(model->phi[i]),cos(model->phi[i]));
+ printf("%4.3f ", model->phi[i]);
+ }
+ }
+ }
+}
+
+
+#define MAX_BINS 40
+//static float bins[] = {2600.0, 2800.0, 3000.0, 3200.0, 3400.0, 3600.0, 3800.0, 4000.0};
+static float bins[] = {/*
+
+ 1000.0, 1100.0, 1200.0, 1300.0, 1400.0,
+ 1500.0, 1600.0, 1700.0, 1800.0, 1900.0,*/
+
+ 2000.0, 2400.0, 2800.0,
+ 3000.0, 3400.0, 3600.0, 4000.0};
+
+void smooth_phase3(struct PEXP *pexp, MODEL *model) {
+ int m, i;
+ int nbins;
+ int b;
+ float f, best_Wo, pred, err;
+ COMP av[MAX_BINS];
+
+ nbins = sizeof(bins)/sizeof(float);
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+
+ /* clear all bins */
+
+ for(i=0; i<MAX_BINS; i++) {
+ av[i].real = 0.0;
+ av[i].imag = 0.0;
+ }
+
+ /* add phases into each bin */
+
+ for(m=1; m<=model->L; m++) {
+ f = m*model->Wo*FS/TWO_PI;
+ if (f > bins[0]) {
+
+ /* find bin */
+
+ for(i=0; i<nbins; i++)
+ if ((f > bins[i]) && (f <= bins[i+1]))
+ b = i;
+ assert(b < MAX_BINS);
+
+ /* est predicted phase from average */
+
+ pred = pexp->phi_prev[m] + N*m*best_Wo;
+ err = model->phi[m] - pred;
+ av[b].real += cos(err); av[b].imag += sin(err);
+ }
+
+ }
+
+ /* use averages to est phases */
+
+ for(m=1; m<=model->L; m++) {
+ f = m*model->Wo*FS/TWO_PI;
+ if (f > bins[0]) {
+
+ /* find bin */
+
+ for(i=0; i<nbins; i++)
+ if ((f > bins[i]) && (f <= bins[i+1]))
+ b = i;
+ assert(b < MAX_BINS);
+
+ /* add predicted phase error to this bin */
+
+ printf("L %d m %d f %4.f b %d\n", model->L, m, f, b);
+
+ pred = pexp->phi_prev[m] + N*m*best_Wo;
+ err = atan2(av[b].imag, av[b].real);
+ printf(" %d: %4.3f -> ", m, model->phi[m]);
+ model->phi[m] = pred + err;
+ model->phi[m] = atan2(sin(model->phi[m]),cos(model->phi[m]));
+ printf("%4.3f\n", model->phi[m]);
+ }
+ }
+ printf("\n");
+}
+
+
+/*
+ Try to code the phase of the largest amplitude in each band. Randomise the
+ phase of the other harmonics. The theory is that only the largest harmonic
+ will be audible.
+*/
+
+void cb_phase1(struct PEXP *pexp, MODEL *model) {
+ int m, i;
+ int nbins;
+ int b;
+ float f, best_Wo;
+ float max_val[MAX_BINS];
+ int max_ind[MAX_BINS];
+
+ nbins = sizeof(bins)/sizeof(float);
+ best_Wo = refine_Wo(pexp, model, 1, model->L);
+
+ for(i=0; i<nbins; i++)
+ max_val[i] = 0.0;
+
+ /* determine max amplitude for each bin */
+
+ for(m=1; m<=model->L; m++) {
+ f = m*model->Wo*FS/TWO_PI;
+ if (f > bins[0]) {
+
+ /* find bin */
+
+ for(i=0; i<nbins; i++)
+ if ((f > bins[i]) && (f <= bins[i+1]))
+ b = i;
+ assert(b < MAX_BINS);
+
+ if (model->A[m] > max_val[b]) {
+ max_val[b] = model->A[m];
+ max_ind[b] = m;
+ }
+ }
+
+ }
+
+ /* randomise phase of other harmonics */
+
+ for(m=1; m<=model->L; m++) {
+ f = m*model->Wo*FS/TWO_PI;
+ if (f > bins[0]) {
+
+ /* find bin */
+
+ for(i=0; i<nbins; i++)
+ if ((f > bins[i]) && (f <= bins[i+1]))
+ b = i;
+ assert(b < MAX_BINS);
+
+ if (m != max_ind[b])
+ model->phi[m] = pexp->phi_prev[m] + N*m*best_Wo;
+ }
+ }
+}
+
+
+/*
+ Theory is only the phase of the envelope of signal matters within a
+ Critical Band. So we estimate the position of an impulse that
+ approximates the envelope of the signal.
+*/
+
+void cb_phase2(struct PEXP *pexp, MODEL *model) {
+ int st, m, i, a, b, step;
+ float diff,w,c,s,phi1_;
+ float A[MAX_AMP];
+
+ for(m=1; m<=model->L; m++) {
+ A[m] = model->A[m];
+ model->A[m] = 0;
+ }
+
+ st = 2*model->L/4;
+ step = 3;
+ model->phi[1] = pexp->phi_prev[1] + (pexp->Wo_prev+model->Wo)*N/2.0;
+
+ printf("L=%d ", model->L);
+ for(m=st; m<st+step*2; m+=step) {
+ a = m; b=a+step;
+ if (b > model->L)
+ b = model->L;
+
+ c = s = 0;
+ for(i=a; i<b-1; i++) {
+ printf("diff %d,%d ", i, i+1);
+ diff = model->phi[i+1] - model->phi[i];
+ //w = (model->A[i+1] + model->A[i])/2;
+ w = 1.0;
+ c += w*cos(diff); s += w*sin(diff);
+ }
+ phi1_ = atan2(s,c);
+ printf("replacing: ");
+ for(i=a; i<b; i++) {
+ //model->phi[i] = i*phi1_;
+ //model->phi[i] = i*model->phi[1];
+ //model->phi[i] = m*(pexp->Wo_prev+model->Wo)*N/2.0;
+ model->A[m] = A[m];
+ printf("%d ", i);
+ }
+ printf(" . ");
+ }
+ printf("\n");
+}
+
+
+static void smooth_phase4(MODEL *model) {
+ int m;
+ float phi_m, phi_m_1;
+
+ if (model->L > 25) {
+ printf("\nL %d\n", model->L);
+ for(m=model->L/2; m<=model->L; m+=2) {
+ if ((m+1) <= model->L) {
+ phi_m = (model->phi[m] - model->phi[m+1])/2.0;
+ phi_m_1 = (model->phi[m+1] - model->phi[m])/2.0;
+ model->phi[m] = phi_m;
+ model->phi[m+1] = phi_m_1;
+ printf("%d %4.3f %4.3f ", m, phi_m, phi_m_1);
+ }
+ }
+ }
+
+}
+
+/* try repeating last frame, just advance phases to account for time shift */
+
+static void repeat_phases(struct PEXP *pexp, MODEL *model) {
+ int m;
+
+ *model = pexp->prev_model;
+ for(m=1; m<=model->L; m++)
+ model->phi[m] += N*m*model->Wo;
+
+}
+
+/*---------------------------------------------------------------------------*\
+
+ phase_experiment()
+
+ Phase quantisation experiments.
+
+\*---------------------------------------------------------------------------*/
+
+void phase_experiment(struct PEXP *pexp, MODEL *model, char *arg) {
+ int m;
+ float before[MAX_AMP], best_Wo;
+
+ assert(pexp != NULL);
+ memcpy(before, &model->phi[0], sizeof(float)*MAX_AMP);
+
+ if (strcmp(arg,"q3") == 0) {
+ quant_phases(model, 1, model->L, 3);
+ update_snr_calc(pexp, model, before);
+ update_variance_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"dec2") == 0) {
+ if ((pexp->frames % 2) != 0) {
+ predict_phases(pexp, model, 1, model->L);
+ update_snr_calc(pexp, model, before);
+ update_variance_calc(pexp, model, before);
+ }
+ }
+
+ if (strcmp(arg,"repeat") == 0) {
+ if ((pexp->frames % 2) != 0) {
+ repeat_phases(pexp, model);
+ update_snr_calc(pexp, model, before);
+ update_variance_calc(pexp, model, before);
+ }
+ }
+
+ if (strcmp(arg,"vq") == 0) {
+ sparse_vq_pred_error(pexp, model);
+ update_snr_calc(pexp, model, before);
+ update_variance_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"pred") == 0)
+ predict_phases_state(pexp, model, 1, model->L);
+
+ if (strcmp(arg,"pred1k") == 0)
+ predict_phases(pexp, model, 1, model->L/4);
+
+ if (strcmp(arg,"smooth") == 0) {
+ smooth_phase(pexp, model,0);
+ update_snr_calc(pexp, model, before);
+ }
+ if (strcmp(arg,"smoothtrain") == 0)
+ smooth_phase(pexp, model,1);
+ if (strcmp(arg,"smoothvq") == 0) {
+ smooth_phase(pexp, model,2);
+ update_snr_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"smooth2") == 0)
+ smooth_phase2(pexp, model);
+ if (strcmp(arg,"smooth3") == 0)
+ smooth_phase3(pexp, model);
+ if (strcmp(arg,"smooth4") == 0)
+ smooth_phase4(model);
+ if (strcmp(arg,"vqsmooth3") == 0) {
+ sparse_vq_pred_error(pexp, model);
+ smooth_phase3(pexp, model);
+ }
+
+ if (strcmp(arg,"cb1") == 0) {
+ cb_phase1(pexp, model);
+ update_snr_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"top") == 0) {
+ //top_amp(pexp, model, 1, model->L/4, 4, 1);
+ //top_amp(pexp, model, model->L/4, model->L/3, 4, 1);
+ //top_amp(pexp, model, model->L/3+1, model->L/2, 4, 1);
+ //top_amp(pexp, model, model->L/2, model->L, 6, 1);
+ //rand_phases(model, model->L/2, 3*model->L/4);
+ //struct_phases(pexp, model, model->L/2, 3*model->L/4);
+ //update_snr_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"pred23") == 0) {
+ predict_phases2(pexp, model, model->L/2, model->L);
+ update_snr_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"struct23") == 0) {
+ struct_phases(pexp, model, model->L/2, 3*model->L/4 );
+ update_snr_calc(pexp, model, before);
+ }
+
+ if (strcmp(arg,"addnoise") == 0) {
+ int m;
+ float max;
+
+ max = 0;
+ for(m=1; m<=model->L; m++)
+ if (model->A[m] > max)
+ max = model->A[m];
+ max = 20.0*log10(max);
+ for(m=1; m<=model->L; m++)
+ if (20.0*log10(model->A[m]) < (max-20)) {
+ model->phi[m] += (PI/4)*(1.0 -2.0*rand()/RAND_MAX);
+ //printf("m %d\n", m);
+ }
+ }
+
+ /* normalise phases */
+
+ for(m=1; m<=model->L; m++)
+ model->phi[m] = atan2(sin(model->phi[m]), cos(model->phi[m]));
+
+ /* update states */
+
+ //best_Wo = refine_Wo(pexp, model, model->L/2, model->L);
+ pexp->phi1 += N*model->Wo;
+
+ for(m=1; m<=model->L; m++)
+ pexp->phi_prev[m] = model->phi[m];
+ pexp->Wo_prev = model->Wo;
+ pexp->frames++;
+ pexp->prev_model = *model;
+}
+
+#ifdef OLD_STUFF
+ //quant_phases(model, 1, model->L, 3);
+ //update_variance_calc(pexp, model, before);
+ //print_sparse_pred_error(pexp, model, 1, model->L, 40.0);
+
+ //sparse_vq_pred_error(pexp, model);
+
+ //quant_phases(model, model->L/4+1, model->L, 3);
+
+ //predict_phases1(pexp, model, 1, model->L/4);
+ //quant_phases(model, model->L/4+1, model->L, 3);
+
+ //quant_phases(model, 1, model->L/8, 3);
+
+ //update_snr_calc(pexp, model, before);
+ //update_variance_calc(pexp, model, before);
+
+ //fixed_bits_per_frame(pexp, model, 40);
+ //struct_phases(pexp, model, 1, model->L/4);
+ //rand_phases(model, 10, model->L);
+ //for(m=1; m<=model->L; m++)
+ // model->A[m] = 0.0;
+ //model->A[model->L/2] = 1000;
+ //repeat_phases(model, 20);
+ //predict_phases(pexp, model, 1, model->L/4);
+ //quant_phases(model, 1, 10, 3);
+ //quant_phases(model, 10, 20, 2);
+ //repeat_phases(model, 20);
+ //rand_phases(model, 3*model->L/4, model->L);
+ // print_phi1_pred_error(model, 1, model->L);
+ //predict_phases(pexp, model, 1, model->L/4);
+ //first_order_band(model, model->L/4, model->L/2);
+ //first_order_band(model, model->L/2, 3*model->L/4);
+ //if (fabs(model->Wo - pexp->Wo_prev)< 0.1*model->Wo)
+
+ //print_pred_error(pexp, model, 1, model->L, 40.0);
+ //print_sparse_pred_error(pexp, model, 1, model->L, 40.0);
+
+ //phi1_est = est_phi1(model, 1, model->L/4);
+ //print_phi1_pred_error(model, 1, model->L/4);
+
+ //first_order_band(model, 1, model->L/4, phi1_est);
+ //sub_linear(model, 1, model->L/4, phi1_est);
+
+ //top_amp(pexp, model, 1, model->L/4, 4);
+ //top_amp(pexp, model, model->L/4, model->L/2, 4);
+
+ //first_order_band(model, 1, model->L/4, phi1_est);
+ //first_order_band(model, model->L/4, model->L/2, phi1_est);
+
+ //if (fabs(model->Wo - pexp->Wo_prev) > 0.2*model->Wo)
+ // rand_phases(model, model->L/2, model->L);
+
+ //top_amp(pexp, model, 1, model->L/4, 4);
+ //top_amp(pexp, model, model->L/4, model->L/2, 8);
+ //top_amp(pexp, model, model->L/4+1, model->L/2, 10, 1);
+ //top_amp(pexp, model, 1, model->L/4, 10, 1);
+ //top_amp(pexp, model, model->L/4+1, 3*model->L/4, 10, 1);
+ //top_amp(pexp, model, 1, 3*model->L/4, 20, 1);
+
+ #ifdef REAS_CAND1
+ predict_phases(pexp, model, 1, model->L/4);
+ top_amp(pexp, model, model->L/4+1, 3*model->L/4, 10, 1);
+ rand_phases(model, 3*model->L/4+1, model->L);
+ #endif
+
+ #ifdef REAS_CAND2
+ if ((pexp->frames % 2) == 0) {
+ //printf("quant\n");
+ predict_phases(pexp, model, 1, model->L/4);
+ //top_amp(pexp, model, model->L/4+1, 3*model->L/4, 20, 1);
+ top_amp(pexp, model, model->L/4+1, 7*model->L/8, 20, 1);
+ rand_phases(model, 7*model->L/8+1, model->L);
+ }
+ else {
+ //printf("predict\n");
+ predict_phases(pexp, model, 1, model->L);
+ }
+ #endif
+
+ //#define REAS_CAND3
+ #ifdef REAS_CAND3
+ if ((pexp->frames % 3) != 0) {
+ printf("pred\n");
+ predict_phases(pexp, model, 1, model->L);
+ }
+ else {
+ predict_phases(pexp, model, 1, model->L/4);
+ fixed_bits_per_frame(pexp, model, model->L/4+1, 60);
+ }
+ #endif
+ //predict_phases(pexp, model, model->L/4, model->L);
+
+
+ //print_pred_error(pexp, model, 1, model->L);
+ //limit_prediction_error(pexp, model, model->L/2, model->L, PI/2);
+#endif
--- /dev/null
+/*---------------------------------------------------------------------------*\
+
+ FILE........: phaseexp.h
+ AUTHOR......: David Rowe
+ DATE CREATED: June 2012
+
+ Experimental functions for quantising, modelling and synthesising phase.
+
+\*---------------------------------------------------------------------------*/
+
+/*
+ Copyright (C) 2012 David Rowe
+
+ All rights reserved.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License version 2.1, as
+ published by the Free Software Foundation. This program is
+ distributed in the hope that it will be useful, but WITHOUT ANY
+ WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+ License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program; if not, see <http://www.gnu.org/licenses/>.
+*/
+
+#ifndef __PHASEEXP__
+#define __PHASEEXP__
+
+#include "kiss_fft.h"
+
+struct PEXP;
+
+struct PEXP * phase_experiment_create();
+void phase_experiment_destroy(struct PEXP *pexp);
+void phase_experiment(struct PEXP *pexp, MODEL *model, char *arg);
+
+#endif
}
}
-/* 3 stage VQ LSP quantsier kindly submitted by Anssi, OH3GDD */
+#define MBEST_STAGES 3
-void lspanssi_quantise(float *x, float *xq, int ndim)
+struct MBEST_LIST {
+ int index[MBEST_STAGES]; /* index of each stage that lead us to this error */
+ float error;
+};
+
+struct MBEST {
+ int entries; /* number of entries in mbest list */
+ struct MBEST_LIST *list;
+};
+
+
+static struct MBEST *mbest_create(int entries) {
+ int i,j;
+ struct MBEST *mbest;
+
+ assert(entries > 0);
+ mbest = (struct MBEST *)malloc(sizeof(struct MBEST));
+ assert(mbest != NULL);
+
+ mbest->entries = entries;
+ mbest->list = (struct MBEST_LIST *)malloc(entries*sizeof(struct MBEST_LIST));
+ assert(mbest->list != NULL);
+
+ for(i=0; i<mbest->entries; i++) {
+ for(j=0; j<MBEST_STAGES; j++)
+ mbest->list[i].index[j] = 0;
+ mbest->list[i].error = 1E32;
+ }
+
+ return mbest;
+}
+
+
+static void mbest_destroy(struct MBEST *mbest) {
+ assert(mbest != NULL);
+ free(mbest->list);
+ free(mbest);
+}
+
+
+/*---------------------------------------------------------------------------*\
+
+ mbest_insert
+
+ Insert the results of a vector to codebook entry comparison. the
+ list is in order of ascending error, so those codebook entries with the
+ smallest error will be first on the list.
+
+\*---------------------------------------------------------------------------*/
+
+static void mbest_insert(struct MBEST *mbest, int index[], float error) {
+ int i, j, found;
+ struct MBEST_LIST *list = mbest->list;
+ int entries = mbest->entries;
+
+ found = 0;
+ for(i=0; i<entries && !found; i++)
+ if (error < list[i].error) {
+ found = 1;
+ for(j=entries-1; j>i; j--)
+ list[j] = list[j-1];
+ for(j=0; j<MBEST_STAGES; j++)
+ list[i].index[j] = index[j];
+ list[i].error = error;
+ }
+}
+
+
+static void mbest_print(char title[], struct MBEST *mbest) {
+ int i,j;
+
+ printf("%s\n", title);
+ for(i=0; i<mbest->entries; i++) {
+ for(j=0; j<MBEST_STAGES; j++)
+ printf(" %4d ", mbest->list[i].index[j]);
+ printf(" %f\n", mbest->list[i].error);
+ }
+}
+
+
+/*---------------------------------------------------------------------------*\
+
+ mbest_search
+
+ Searches vec[] to a codebbook of vectors, and maintains a list of the mbest
+ closest matches.
+
+\*---------------------------------------------------------------------------*/
+
+static void mbest_search(
+ const float *cb, /* VQ codebook to search */
+ float vec[], /* target vector */
+ float w[], /* weighting vector */
+ int k, /* dimension of vector */
+ int m, /* number on entries in codebook */
+ struct MBEST *mbest, /* list of closest matches */
+ int index[] /* indexes that lead us here */
+)
{
- int i, n1, n2, n3;
- float e, err1[LPC_ORD], err2[LPC_ORD], err3[LPC_ORD];
+ float e;
+ int i,j;
+ float diff;
+
+ for(j=0; j<m; j++) {
+ e = 0.0;
+ for(i=0; i<k; i++) {
+ diff = cb[j*k+i]-vec[i];
+ e += pow(diff*w[i],2.0);
+ }
+ index[0] = j;
+ mbest_insert(mbest, index, e);
+ }
+}
+
+
+/* 3 stage VQ LSP quantiser. Design and guidance kindly submitted by Anssi, OH3GDD */
+
+void lspanssi_quantise(float *x, float *xq, int ndim, int mbest_entries)
+{
+ int i, j, n1, n2, n3;
float w[LPC_ORD];
const float *codebook1 = lsp_cbvqanssi[0].cb;
const float *codebook2 = lsp_cbvqanssi[1].cb;
const float *codebook3 = lsp_cbvqanssi[2].cb;
-
+ struct MBEST *mbest_stage1, *mbest_stage2, *mbest_stage3;
+ float target[LPC_ORD];
+ int index[MBEST_STAGES];
+
+ mbest_stage1 = mbest_create(mbest_entries);
+ mbest_stage2 = mbest_create(mbest_entries);
+ mbest_stage3 = mbest_create(mbest_entries);
+ for(i=0; i<MBEST_STAGES; i++)
+ index[i] = 0;
+
compute_weights_anssi(x, w, ndim);
+
#ifdef DUMP
dump_weights(w, ndim);
#endif
- n1 = find_nearest_weighted(codebook1, lsp_cbvqanssi[0].m, x, w, ndim);
-
- e = 0.0;
- for (i=0;i<ndim;i++)
- {
- xq[i] = codebook1[ndim*n1+i];
- err1[i] = x[i] - xq[i];
- e += err1[i]*err1[i];
+ /* Stage 1 */
+
+ mbest_search(codebook1, x, w, ndim, lsp_cbvqanssi[0].m, mbest_stage1, index);
+ mbest_print("Stage 1:", mbest_stage1);
+
+ /* Stage 2 */
+
+ for (j=0; j<mbest_entries; j++) {
+ index[1] = n1 = mbest_stage1->list[j].index[0];
+ for(i=0; i<ndim; i++)
+ target[i] = x[i] - codebook1[ndim*n1+i];
+ mbest_search(codebook2, target, w, ndim, lsp_cbvqanssi[1].m, mbest_stage2, index);
}
- //printf("error1: %f\n", e);
+ mbest_print("Stage 2:", mbest_stage2);
- n2 = find_nearest_weighted(codebook2, lsp_cbvqanssi[1].m, err1, w, ndim);
+ /* Stage 3 */
- e = 0.0;
- for (i=0;i<ndim;i++)
- {
- xq[i] += codebook2[ndim*n2+i];
- err2[i] = x[i] - xq[i];
- e += err2[i]*err2[i];
+ for (j=0; j<mbest_entries; j++) {
+ index[2] = n1 = mbest_stage2->list[j].index[1];
+ index[1] = n2 = mbest_stage2->list[j].index[0];
+ for(i=0; i<ndim; i++)
+ target[i] = x[i] - codebook1[ndim*n1+i] - codebook2[ndim*n2+i];
+ mbest_search(codebook3, target, w, ndim, lsp_cbvqanssi[2].m, mbest_stage3, index);
}
- //printf("error2: %f\n", e);
+ mbest_print("Stage 3:", mbest_stage3);
- n3 = find_nearest_weighted(codebook3, lsp_cbvqanssi[2].m, err2, w, ndim);
-
- e = 0.0;
+ n1 = mbest_stage3->list[0].index[2];
+ n2 = mbest_stage3->list[0].index[1];
+ n3 = mbest_stage3->list[0].index[0];
for (i=0;i<ndim;i++)
- {
- xq[i] += codebook3[ndim*n3+i];
- err3[i] = x[i] - xq[i];
- e += err3[i]*err3[i];
- }
- //printf("error3: %f\n", e);
+ xq[i] = codebook1[ndim*n1+i] + codebook2[ndim*n2+i] + codebook3[ndim*n3+i];
+
+ mbest_destroy(mbest_stage1);
+ mbest_destroy(mbest_stage2);
+ mbest_destroy(mbest_stage3);
}
int check_lsp_order(float lsp[], int lpc_order)
void lspjnd_quantise(float lsp[], float lsp_[], int order);
void lspdt_quantise(float lsps[], float lsps_[], float lsps__prev[], int mode);
void lspjvm_quantise(float lsps[], float lsps_[], int order);
-void lspanssi_quantise(float lsps[], float lsps_[], int order);
+void lspanssi_quantise(float lsps[], float lsps_[], int order, int mbest_entries);
void quantise_WoE(MODEL *model, float *e, float xq[]);
int encode_WoE(MODEL *model, float e, float xq[]);
int quantise(float cb[], float vec[], int d, int e, float *se);
void print_vec(float cb[], int d, int e);
void split(float cb[], int d, int b);
+int gain_shape_quantise(float cb[], float vec[], int d, int e, float *se, float *best_gain);
/*-----------------------------------------------------------------------* \
/* Interpret command line arguments */
- if (argc != 5) {
- printf("usage: %s TrainFile D(dimension) B(number of bits) VQFile\n", argv[0]);
+ if (argc < 5) {
+ printf("usage: %s TrainFile D(dimension) B(number of bits) VQFile [error.txt file]\n", argv[0]);
exit(1);
}
/* main loop */
printf("\n");
- printf("bits Iteration delta var std dev\n");
- printf("--------------------------------------\n");
+ printf("bits Iteration delta var std dev\n");
+ printf("---------------------------------------\n");
for(b=1; b<=bits; b++) {
levels = 1<<b;
for(i=0; i<J; i++) {
ret = fread(vec, sizeof(float), d, ftrain);
ind = quantise(cb, vec, d, levels, &se);
+ //ind = gain_shape_quantise(cb, vec, d, levels, &se, &best_gain);
+ //for(j=0; j<d; j++)
+ // if (vec[j] != 0.0)
+ // vec[j] += best_gain;
#ifdef DBG
print_vec(vec, d, 1);
printf(" ind %d se: %f\n", ind, se);
finished = 1;
}
- /* determine new codebook from centroids */
+ if (!finished) {
+ /* determine new codebook from centroids */
- for(i=0; i<levels; i++) {
- norm(¢[i*d], d, &n[i*d]);
- memcpy(&cb[i*d], ¢[i*d], d*sizeof(float));
+ for(i=0; i<levels; i++) {
+ norm(¢[i*d], d, &n[i*d]);
+ memcpy(&cb[i*d], ¢[i*d], d*sizeof(float));
+ }
}
-
+
#ifdef DBG
printf("new cb ...\n");
print_vec(cent, d, e);
}
fclose(fvq);
+ /* optionally dump error file for multi-stage work */
+
+ if (argc == 6) {
+ FILE *ferr = fopen(argv[5],"wt");
+ assert(ferr != NULL);
+ rewind(ftrain);
+ for(i=0; i<J; i++) {
+ ret = fread(vec, sizeof(float), d, ftrain);
+ ind = quantise(cb, vec, d, levels, &se);
+ for(j=0; j<d; j++) {
+ if (vec[j] != 0.0)
+ vec[j] -= cb[ind*d+j];
+ fprintf(ferr, "%f ", vec[j]);
+ }
+ fprintf(ferr, "\n");
+ }
+ }
+
return 0;
}
return(besti);
}
+int gain_shape_quantise(float cb[], float vec[], int d, int e, float *se, float *best_gain)
+{
+ float error; /* current error */
+ int besti; /* best index so far */
+ float best_error; /* best error so far */
+ int i,j,m;
+ float diff, metric, best_metric, gain, sumAm, sumCb;
+
+ besti = 0;
+ best_metric = best_error = 1E32;
+ for(j=0; j<e; j++) {
+
+ /* compute optimum gain */
+
+ sumAm = sumCb = 0.0;
+ m = 0;
+ for(i=0; i<d; i++) {
+ if (vec[i] != 0.0) {
+ m++;
+ sumAm += vec[i];
+ sumCb += cb[j*d+i];
+ }
+ }
+ gain = (sumAm - sumCb)/m;
+
+ /* compute error */
+
+ metric = error = 0.0;
+ for(i=0; i<d; i++) {
+ if (vec[i] != 0.0) {
+ diff = vec[i] - cb[j*d+i] - gain;
+ error += diff*diff;
+ metric += diff*diff;
+ }
+ }
+ if (metric < best_metric) {
+ best_error = error;
+ best_metric = metric;
+ *best_gain = gain;
+ besti = j;
+ }
+ }
+
+ *se += best_error;
+
+ return(besti);
+}
+
void split(float cb[], int d, int levels)
{
int i,j;
projects / freetel-svn-tracking.git / commitdiff