freeswitch/libs/libcodec2/unittest
Anthony Minessale efd3744a21 FS-7122: Working with william to find a better way 2015-02-16 16:25:35 -06:00
..
Makefile.am FS-7122: Working with william to find a better way 2015-02-16 16:25:35 -06:00
README
create_interleaver.c
de.c
dvdongle2.c
extract.c
ge_train.c
genampdata.c
genlsp.c
genphdata.c
genres.c
lsp1.txt
lsp2.txt
lsp3.txt
lsp4.txt
lsp5.txt
lsp6.txt
lsp7.txt
lsp8.txt
lsp9.txt
lsp10.txt
lsp45678910.txt
lspd78.txt
lspd123.txt
lspd456.txt
lspd910.txt
lspd678910.txt
lspjnd5-10.txt
lspsync.c
lsptest.c
mksine.c
polar2rect.c
pre.c
run_tests.sh
scalarlsptest.c
sd.c
sd.h
speexlsptest.c
t48_8.c
tcodec2.c
tcontphase.c
tfdmdv.c
tfifo.c
tinterp.c
tlspsens.c
tnlp.c
tprede.c
tquant.c
tsrc.c
vq_train_jvm.c
vqtrain.c
vqtrainjnd.c
vqtrainph.c
vqtrainsp.c

README

README
for codec2/unittest
Created David Rowe 31 July 2012

Training (experimental) sparse phase VQs:

1/ In ../src/phase.c phase_experiment() enable:

   print_sparse_pred_error()

   and 'make' c2sim (in src)

2/ Run over a training database:

  $ ./c2sim /xhome1/codec2/samples/train.spc --phaseexp > train_phtrain.txt

  a) check stats in Octave:
     octave> load ../src/train_phtrain.txt
     octave> std(nonzeros(train_phtrain(:,1:20)))
     octave> hist(nonzeros(train_phtrain(:,1:20)),20)
3/ Extract and convert to floats vector you wish to train for example
   first 20 (out of MAX_AMP == 80):

  $ ./extract ../src/train_phtrain.txt train_phtrain.flt 1 20

4/ Convert to rectangular:

  $ ./polar2rect train_phtrain.flt train_phtrainr.flt

5/ Run this program:

  $ ./vqtrainph train_phtrainr.flt 20 1024 vq.txt

  Ouput is vq.txt

Tests
-----

+ build up insmallest possible stesp
+ impl errors v alg errors
+ use actual phase data as codebook
+ test vq with rand phases first or known data