71e046b0ba
E2E/Streaming Transformer/Conformer ASR ( #578 )
...
* add cmvn and label smoothing loss layer
* add layer for transformer
* add glu and conformer conv
* add torch compatiable hack, mask funcs
* not hack size since it exists
* add test; attention
* add attention, common utils, hack paddle
* add audio utils
* conformer batch padding mask bug fix #223
* fix typo, python infer fix rnn mem opt name error and batchnorm1d, will be available at 2.0.2
* fix ci
* fix ci
* add encoder
* refactor egs
* add decoder
* refactor ctc, add ctc align, refactor ckpt, add warmup lr scheduler, cmvn utils
* refactor docs
* add fix
* fix readme
* fix bugs, refactor collator, add pad_sequence, fix ckpt bugs
* fix docstring
* refactor data feed order
* add u2 model
* refactor cmvn, test
* add utils
* add u2 config
* fix bugs
* fix bugs
* fix autograd maybe has problem when using inplace operation
* refactor data, build vocab; add format data
* fix text featurizer
* refactor build vocab
* add fbank, refactor feature of speech
* refactor audio feat
* refactor data preprare
* refactor data
* model init from config
* add u2 bins
* flake8
* can train
* fix bugs, add coverage, add scripts
* test can run
* fix data
* speed perturb with sox
* add spec aug
* fix for train
* fix train logitc
* fix logger
* log valid loss, time dataset process
* using np for speed perturb, remove some debug log of grad clip
* fix logger
* fix build vocab
* fix logger name
* using module logger as default
* fix
* fix install
* reorder imports
* fix board logger
* fix logger
* kaldi fbank and mfcc
* fix cmvn and print prarams
* fix add_eos_sos and cmvn
* fix cmvn compute
* fix logger and cmvn
* fix subsampling, label smoothing loss, remove useless
* add notebook test
* fix log
* fix tb logger
* multi gpu valid
* fix log
* fix log
* fix config
* fix compute cmvn, need paddle 2.1
* add cmvn notebook
* fix layer tools
* fix compute cmvn
* add rtf
* fix decoding
* fix layer tools
* fix log, add avg script
* more avg and test info
* fix dataset pickle problem; using 2.1 paddle; num_workers can > 0; ckpt save in exp dir;fix setup.sh;
* add vimrc
* refactor tiny script, add transformer and stream conf
* spm demo; librisppech scripts and confs
* fix log
* add librispeech scripts
* refactor data pipe; fix conf; fix u2 default params
* fix bugs
* refactor aishell scripts
* fix test
* fix cmvn
* fix s0 scripts
* fix ds2 scripts and bugs
* fix dev & test dataset filter
* fix dataset filter
* filter dev
* fix ckpt path
* filter test, since librispeech will cause OOM, but all test wer will be worse, since mismatch train with test
* add comment
* add syllable doc
* fix ds2 configs
* add doc
* add pypinyin tools
* fix decoder using blank_id=0
* mmseg with pybind11
* format code
3 years ago
e0a87a5ab1
batch average ctc loss ( #567 )
...
* when loss div batchsize, change lr, more epoch, loss can reduce more and cer lower than before
* since loss reduce more when loss div batchsize, less lm alpha can be better.
* less lm alpha, more cer reduce
* alpha 2.2, cer 0.077478
* alpha 1.9, cer 0.077249
* large librispeech lr for batch_average ctc loss
* since loss reduce and model more confidence, then less lm alpha
3 years ago
258307df9b
fix egs bugs ( #552 )
...
* fix egs
* fix log
3 years ago
1539f3e0a3
Refactor CTC module, add embedding and fix log ( #549 )
...
* add acts, refactor ctc, add pos embed
* fix export, dataloader time log
* fix egs
* fix libri readme
3 years ago
19e0f2ac46
Fix doc format ( #546 )
3 years ago
57ed5cd2e0
Fix Doc ( #544 )
3 years ago
d7e753546a
Support paddle 2.x ( #538 )
...
* 2.x model
* model test pass
* fix data
* fix soundfile with flac support
* one thread dataloader test pass
* export feasture size
add trainer and utils
add setup model and dataloader
update travis using Bionic dist
* add venv; test under venv
* fix unittest; train and valid
* add train and config
* add config and train script
* fix ctc cuda memcopy error
* fix imports
* fix train valid log
* fix dataset batch shuffle shift start from 1
fix rank_zero_only decreator error
close tensorboard when train over
add decoding config and code
* test process can run
* test with decoding
* test and infer with decoding
* fix infer
* fix ctc loss
lr schedule
sortagrad
logger
* aishell egs
* refactor train
add aishell egs
* fix dataset batch shuffle and add batch sampler log
print model parameter
* fix model and ctc
* sequence_mask make all inputs zeros, which cause grad be zero, this is a bug of LessThanOp
add grad clip by global norm
add model train test notebook
* ctc loss
remove run prefix
using ord value as text id
* using unk when training
compute_loss need text ids
ord id using in test mode, which compute wer/cer
* fix tester
* add lr_deacy
refactor code
* fix tools
* fix ci
add tune
fix gru model bugs
add dataset and model test
* fix decoding
* refactor repo
fix decoding
* fix musan and rir dataset
* refactor io, loss, conv, rnn, gradclip, model, utils
* fix ci and import
* refactor model
add export jit model
* add deploy bin and test it
* rm uselss egs
* add layer tools
* refactor socket server
new model from pretrain
* remve useless
* fix instability loss and grad nan or inf for librispeech training
* fix sampler
* fix libri train.sh
* fix doc
* add license on cpp
* fix doc
* fix libri script
* fix install
* clip 5 wer 7.39, clip 400 wer 7.54, 1.8 clip 400 baseline 7.49
3 years ago
141109b49d
update aishell egs
3 years ago
7357e52188
fix egs
3 years ago
9cfca9fc40
fix dataset dir
3 years ago
123d7a6f3f
refactor egs
3 years ago
ba7cf0782e
add copyright
3 years ago
75c8018eab
refactor tiny egs
3 years ago
126677a35c
support py3
3 years ago
b4d3eeadda
change the default configuration in infer.py and test.py
5 years ago
d89c3a48a7
unify api to 1.6 version and fix some problems
5 years ago
d74f4ff3f5
update deepspeech to fluid api
5 years ago
3ecf1ad4f5
Decouple ext scorer init & inference & decoding for the convenience of tuning
7 years ago
6e4f246517
retune hyper-parameters for librispeech model
7 years ago
c786b18265
Uncomment model downloading lines
7 years ago
938685baa0
Add infer & test script for baidu_en8k model
7 years ago
24cb1866a8
Adapt settings for Aishell example.
7 years ago
9800d14956
fix the data path in the librispeech example
7 years ago
528895cf26
fix the model path in the tiny example
7 years ago
edaed68f33
Involve dev dataset to build vocabulary in the aishell example
7 years ago
d58dd5fe72
Release BaiduEn8k model and prepare demo example.
7 years ago
ad08dd963a
supply param cutoff_top_n in several scripts of DS2
7 years ago
9ac15f8bd8
Merge pull request #363 from xinghai-sun/default_value
...
Reset default value of batch_size, num_proc_data and fix an invalid url for DS2.
7 years ago
86811af7a3
Reset default value of batch_size, nun_proc_data and fix an invalid url for DS2.
7 years ago
758e9887ca
Bug fix for librispeech/run_train.sh.
7 years ago
be1fbc68a1
Set process daemon property and reset default value of num_proc_data arguments.
7 years ago
2c63f3413f
convert decoding results to unicode in DS2
7 years ago
8dcf1d5667
fix bugs in preparing data scripts
7 years ago
3663ba57d0
make *.sh files more robust in DS2
7 years ago
6cac602bc5
Merge pull request #194 from kuke/improve_tuning_dev
...
Improve params tuning strategy for CTC beam search decoder
7 years ago
e6e7b13222
remove the log praser in tuning script
7 years ago
4c64841c16
add tuning script & enable ploting error surface
7 years ago
0a5f123764
Add inference scripts.
7 years ago
a235b3532d
Add test script.
7 years ago
7b9385899b
Add training script.
7 years ago
e9a42044f5
Add data preparing for Aishell.
7 years ago
3615b44ffd
Merge branch 'develop' of upstream into ctc_decoder_deploy
7 years ago
7f45752a13
Add profile.sh script for multi-gpu profiling.
7 years ago
acf2072ed4
Merge branch 'develop' of upstream into ctc_decoder_deploy
7 years ago
1b206b3390
fix bugs for model.py and demo_server.py.
7 years ago
a24d0138d9
adjust scorer's init & add logging for scorer & separate long functions
7 years ago
7e093ed1a3
expose param cutoff_top_n
7 years ago
a18e6a7eda
refine by following review comments
7 years ago
cd635cf6f3
Release librispeech model url.
7 years ago
f2120bb5dc
adapt to the new structure
7 years ago
Hui Zhang
lfchener
Yibing Liu
yangyaming
Xinghai Sun
Yibing Liu