A. Kunikoshi p280427
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-04-22 02:04:03 +02:00
97486e5599 dataset for experiments in check_novoapi is updated.
2004399179 novoapi_functions.py is adjusted to use convert_phoneset.py.
Compare 2 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-03-25 00:07:07 +01:00
b444b70af9 fame_phonetics.py and functions to make quests.hed to tie triphone are added.
bf586fcde5 triphone training is added.
fdd165ce6a re-aligned mlf file include less files than original mlf file. Therefore the scp file should also be updated accordingly, when re-estimation is performed. this bug is fixed.
Compare 3 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-03-07 22:17:01 +01:00
fa81b70b27 monophone training is completed.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-03-05 00:11:55 +01:00
41d4fa5ff9 sp is added to the model.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-03-03 02:05:51 +01:00
b1b1942fa0 test on stimmen data is added.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-02-14 00:21:47 +01:00
c185072d5b label alignment using HVite is added.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-02-08 14:10:47 +01:00
8f89f60538 dataset is made.
f6e563ecd3 moved testing parts in htk_vs_kaldi into stimmen_test.py
da0242b0e1 make sure all the phones in stimmen transcription can be treated correctly.
Compare 3 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-02-04 20:33:04 +01:00
ab3887c6ca sp is added to the model.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-02-04 13:46:39 +01:00
f6e7c8eefa bug related encoding on label file is fixed.
322a8a0079 label files are extracted. hcompv_scp is made.
22cccfb61d fix the bug there are characters in the lexicon which cannot be described in ascii.
dc6b7b84b6 lexicon is made.
8cda93de75 fame_asr phoneset is added including reduced version and htk compatible version.
Compare 5 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-27 23:52:52 +01:00
87abbbb95a correspondence between lex.asr and lex.ipa is automatically obtained. header is added to the functions in fame_functions.py.
813f013d7a phonset is given as fame_phoneset.py. translation key is obtained based on the information.
Compare 2 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-24 09:45:51 +01:00
7844a56281 HTK related functions are moved to pyhtk project. fame acoustic models are made using fame_hmm.py. feature extraction is completed. A function is being made to get translation key from ipa to asr.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-21 21:58:13 +01:00
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-21 21:57:25 +01:00
24ac56ac0e to transfer working environment to McRoberts laptop.
82a8e2302f rozen-test is added.
Compare 2 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-20 13:49:11 +01:00
de5c9cecb9 The bug regarding novoapi for Python 3.6 is solved. The detail can be found in novoapi_for_python3x/readme.txt
8efb091715 confusion matrix is output.
05e8a671c1 forced alignment by novoapi is performed.
Compare 3 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-12 23:30:08 +01:00
6edde06a4f check frequency of the pronunciation variants of each word.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-10 23:39:14 +01:00
1622655542 functions are added to perform forced_alignment using novoapi. results can be written in novo70 or IPA.
A. Kunikoshi pushed to master at p280427/acoustic_model 2019-01-07 23:27:16 +01:00
d6d5543d03 novoapi_functions is added to novo70 specific functions.
d6e005b1cb find pronunciation variants which all phones are in novo70.
dd9e3d820b started to check which words in stimmen transcription consists of only phones in novo70 phoneset.
Compare 3 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2018-12-30 23:48:05 +01:00
e5cf182a18 novo_api for python 3.x is added.
A. Kunikoshi pushed to master at p280427/acoustic_model 2018-12-30 23:34:34 +01:00
a77ed9d4dd resolve the conflicts.
df046ffc26 forced_alignment_novo.py is removed (.gitignore). commit history is cleaned up.
Compare 2 commits »
A. Kunikoshi pushed to master at p280427/acoustic_model 2018-12-30 23:19:27 +01:00
beff33fdf9 revert "load novo phoneset" to continue working.
af785e51cf commit to clean up
Compare 2 commits »