d1f5123bcf
Merge pull request #99 from PaddlePaddle/new_augmentation_2
...
Add some audio transformation methods to AudioSegment and SpeechSegment class.
8 years ago
def66a3223
Follow comments.
8 years ago
ddb2bdc190
add audio file
8 years ago
25ce7ebe7b
add audio file
8 years ago
36743d3689
add scoring last word in beam search
8 years ago
21161b0165
add audio file
8 years ago
107f8b89ae
add audio augmentation
8 years ago
ff01d048d3
final refining on old data provider: enable pruning & add evaluation & code cleanup
8 years ago
b8341da63d
add audio augmentation
8 years ago
26eb54eb37
Follow comments.
8 years ago
86eeb52c06
Merge branch 'develop' of https://github.com/PaddlePaddle/models into fix-81
8 years ago
5ca270d30a
add audio file
8 years ago
d1ee10be10
modify audio and speech
8 years ago
d66d740ea0
add audio part
8 years ago
ed5f04afb8
Add shuffle type of instance_shuffle and batch_shuffle_clipped.
8 years ago
65e34c535b
add augmentation
8 years ago
04a225ae4f
Enable min_batch_num in train.py and update train info print.
8 years ago
f85f8558cf
Add unittest.
8 years ago
397b2fc288
Merge branch 'develop' of https://github.com/PaddlePaddle/models into fix-81
8 years ago
1cef98f210
Update README.md for DS2.
8 years ago
b07ee84a1d
Add function, class and module docs for data parts in DS2.
8 years ago
a633eb9cc6
change two arguments
8 years ago
7db13ca9db
enable lm in multiprocessing decoder & add script for params tuning
8 years ago
bc3224eb14
Follow comments.
8 years ago
cd3617aeb4
Refactor whole data preprocessor for DS2 (re-design classes, re-organize dir, add augmentaion interfaces etc.).
...
1. Refactor data preprocessor with new added class AudioSegment, SpeechSegment, TextFeaturizer, AudioFeaturizer, SpeechFeaturizer.
2. Add data augmentation interfaces and class AugmentorBase, AugmentationPipeline, VolumnPerturbAugmentor etc..
3. Seperate normalizer's mean and std computing from training, by adding FeatureNormalizer and a seperate tool compute_mean_std.py.
4. Re-organize directory.
8 years ago
9c27b1d14e
add more comments and update train.py
8 years ago
bf73540067
Merge branch 'develop' of https://github.com/PaddlePaddle/models into ds2
8 years ago
c25c62b8f9
refine audio_data_utils.py
8 years ago
bb34e90398
correct typos in annotations
8 years ago
ae83a25aff
add beam search decoder using multiprocesses
8 years ago
453f038df9
optimize the efficiency of beam search
8 years ago
9fda521ee3
improve external scorer
8 years ago
b046e651e7
tiny modify to pass CI
8 years ago
44efbed798
rename variables in decoder
8 years ago
21ff590e6d
modify language model scoring
8 years ago
ac370eca85
add annotations
8 years ago
51f35a5372
code clean & add external scorer
8 years ago
dedbfb2654
enable ctc beam search decoder
8 years ago
cfe9d22866
update annotations
8 years ago
c943ca79ac
mv ctc_beam_search_decoder into deep_speech_2/
8 years ago
e016778e20
Merge pull request #78 from xinghai-sun/ds2_libri
...
Update librispeech.py for DeepSpeech2.
8 years ago
06e9f71389
Remove manifest's line number check from librispeech.py and update README.md.
8 years ago
3f63e069e0
Fix typos and follow comments.
8 years ago
cc2a4d4e3d
Add error rate calculation script.
8 years ago
f49eab5fec
Change assert to exception raising.
8 years ago
d3eeb7fd76
Refine librispeech.py for DeepSpeech2.
...
Summary:
1. Add manifest line check.
2. Avoid re-unpacking if unpacked data already exists.
3. Add full_download (download all 7 sub-datasets of LibriSpeech).
8 years ago
7c85e0fdb5
Support variable input batch and sortagrad.
8 years ago
d2e467385d
Add loading model function for train.py.
8 years ago
730d5c4dd3
Update DS2 README.md and fix bug in librispeech.py
8 years ago
2a83486500
Refactor decoder interfaces and add ./data directory.
8 years ago
Xinghai Sun
yangyaming
chrisxu2016
Yibing Liu
dangqingqing