b3ebf3fd62
Support padding removing.
7 years ago
a084394128
Add doc and adjust some codes.
7 years ago
39dbcb4dfb
Give option to disable converting from transcription text to ids.
7 years ago
86811af7a3
Reset default value of batch_size, nun_proc_data and fix an invalid url for DS2.
7 years ago
2e5e9b8c11
Turn on rnn_use_batch of Paddle for accelartion.
...
Improve xmap_reader_mp by adding a flush thread.
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
f2120bb5dc
adapt to the new structure
7 years ago
ac56a2f249
Update READMD.md and other details by following reviewers comments.
7 years ago
ae7ef7929a
Rename some folders and update examples.
7 years ago
7d0458c7f7
adapt to the new folder structure of DS2
7 years ago
0bbb9c3ee2
Re-organize folder structure and hierarchy for DS2.
7 years ago
e8f7a8fde1
Update argument naming following Yibing's reviews.
7 years ago
9571b6fc0e
Add back utils.py.
7 years ago
8b64ef29c8
Re-style the config codes for tools in DS2.
7 years ago
792129166a
Sort the config lines to make it look better.
7 years ago
805846ce67
Reduce the config parsing codes for DS2 and make it looks cleaner.
7 years ago
2aa4af1c29
Revert back to support input-hidden weights sharing between bi-directional RNNs.
...
1. Add options to enable and disable RNN weights sharing.
2. Set rnn_layer_size to 2048 by default.
3. Revert back the striding steps of 1st conv layer from 2 to 3.
4. Revert back to BRelu.
Above follows DS2 papers.
7 years ago
177af05953
Fix a bug in use_gru argument parsing.
7 years ago
5a63275845
Add GRU support.
7 years ago
4b3f768df7
Simplify description and codes.
7 years ago
5ef300f3f0
Make type of error rate optional.
7 years ago
6df0f9bc44
Reset default multi-thread/process number to half of cpu count() for speedup.
7 years ago
0ebf36b98f
Add a realtime ASR demo for users to test their own voice with mic.
8 years ago
526e18b119
Add function docs for layer.py and model.py and update other details.
8 years ago
8122dd9c29
Simplify train.py, evaluate.py, infer.py and tune.py by adding DeepSpeech2Model class.
8 years ago
92eacf548b
Update default config params and result display for evaluator.py and infer.py for DS2.
8 years ago
724ef18596
update several scripts to support mfcc
8 years ago
d15c48d616
upload the language model
8 years ago
0cadc56a84
follow comments in code format
8 years ago
8ba98cb518
fix decoders' unittest
8 years ago
26510f74a6
refine ctc_beam_search_decoder
8 years ago
7324d41e4a
Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev
8 years ago
13f708739b
Improve audio featurizer and add shift augmentor.
...
1. Improve audio featurizer.
2. Add shift augmentor.
3. Update default argument to be the current best seggestion.
4. Add checkpoints with pass id.
8 years ago
8033845615
add unit test for decoders
8 years ago
f20c6d643f
Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev
8 years ago
d104eccf67
Update the default num_threads for DS2 data generator.
8 years ago
1d8cc4a5a9
Add multi-threading support for DS2 data generator.
8 years ago
0729abe02e
tiny adjust
8 years ago
b2cf96cc46
adapt to the new data provider
8 years ago
ff01d048d3
final refining on old data provider: enable pruning & add evaluation & code cleanup
8 years ago
ed5f04afb8
Add shuffle type of instance_shuffle and batch_shuffle_clipped.
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
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
bb34e90398
correct typos in annotations
8 years ago
ae83a25aff
add beam search decoder using multiprocesses
8 years ago
ac370eca85
add annotations
8 years ago
yangyaming
Xinghai Sun
Yibing Liu