mirrors/DeepSpeech
2
0
Fork 0
mirror of https://github.com/mozilla/DeepSpeech.git synced 2025-10-26 11:19:39 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,541 Commits 63 Branches 109 Tags 17 GiB
969b2ac4ba
Commit Graph

236 Commits

Author SHA1 Message Date
Tilman Kamp
6b1d6773de SDB support 2020-03-10 10:32:58 +01:00
Daniel
17ddc5600e Reduce learning rate on plateau. 2020-02-18 16:17:51 +01:00
Reuben Morais
bd8b96c19d Remove unneeded Saver instances 2020-02-17 15:55:16 +01:00
Reuben Morais
0e2f34f8cf Synchronize TensorFlow logging with --log_level flag 2020-02-17 12:46:00 +01:00
Reuben Morais
c46d8396bc Respect --load when exporting 2020-02-17 12:45:37 +01:00
JRMeyer
5bba9ea5d1 Transfer-learning support 2020-02-17 08:29:10 +01:00
Reuben Morais
3637f88c06 Fix CI errors, address comments, update examples 2020-02-12 10:13:02 +01:00
Reuben Morais
c512383aec Fix consumers of DS_CreateModel 2020-02-12 10:13:02 +01:00
Reuben Morais
8e9b6ef7b3 Embed default beam width into exported graph and remove param from DS_CreateModel 2020-02-11 21:23:05 +01:00
Reuben Morais
1e2eb96248 Update all API consumers 2020-02-11 19:44:36 +01:00
Alexandre Lissy
ff401732a3 Enforce CTC decoder version check 
Fix #2710
 2020-02-03 10:24:21 +01:00
Reuben Morais
50830d7022 Fix whitespace 2020-01-27 15:59:39 +01:00
PedroDKE
3e349497ed added an argument to choose the final export model name 2020-01-25 12:30:07 +01:00
Reuben Morais
f6cd28ba2d
Fix handling of InvalidArgumentError in training loop 
X-DeepSpeech: NOBUILD
 2020-01-08 17:10:47 +00:00
Reuben Morais
e1d14eb9a9
Merge pull request #2560 from mychiux413/sparse_warp 
[SpecAugment] Refactor sparse_image_warp for dynamic shape of spectrogram
 2020-01-03 13:25:31 +01:00
Yi-Hua Chiu
4133e620bd remove debugging leftover, add UNSUPPORTED note, just skip invertible error 2020-01-03 10:32:02 +08:00
Yi-Hua Chiu
c570cb670a sparse_warp still can have chance to raise error even after millions steps, so just recover the invertible error while training, unless error raise 3 times, it will be aborted 2020-01-02 11:06:15 +08:00
Alexandre Lissy
5f003cfbd6 Set forget_bias=0 for static RNN implementation 
Fixes #2612
 2019-12-20 13:28:46 +01:00
Yi-Hua Chiu
5fbc7e8596 [FIX] use time_warping_para to constraint control width, add logic to disable cache [ADD] num_control_points 2019-12-04 10:25:36 +08:00
Reuben Morais
e3b1b5fd42 Disable caching features to memory 2019-11-29 10:17:15 +01:00
Tilman Kamp
a160dc72c4 Fix: Added executable flag to DeepSpeech.py again 2019-11-21 13:08:53 +01:00
Tilman Kamp
29528ed7b7 Separate process per file; less log noise 2019-11-20 17:29:13 +01:00
Alexandre Lissy
fe6230020c Support packaging as Zip file 2019-11-13 11:16:11 +01:00
Reuben Morais
8c82081779 Embed alphabet directly in model 2019-11-05 09:02:21 +01:00
Reuben Morais
31ec7a71f2 Fix undefined variable when saving test samples to --test_output_file 
X-DeepSpeech: NOBUILD
 2019-10-30 10:30:00 +01:00
Reuben Morais
44a605c8b7
Merge pull request #2435 from mozilla/uplift-utf8-fixes 
Uplift general fixes from UTF-8 work
 2019-10-25 09:09:48 +00:00
Reuben Morais
12baf5ffbc Expose cutoff_prob and cutoff_top_n as flags 2019-10-23 11:15:23 +02:00
Alexandre Lissy
1939f74ec0 Store graph version on TFLite 2019-10-16 15:36:45 +02:00
Reuben Morais
ef3bdb2540 Merge PR #2434 - Add flag for automatic mixed precision training 2019-10-15 13:45:19 +02:00
Reuben Morais
31922cb3dc Clarify docs and fix linter 2019-10-15 13:44:09 +02:00
Reuben Morais
e0d8ef75e8 Use try_loading and FLAGS.load in --one_shot_infer code 2019-10-14 21:03:14 +02:00
Reuben Morais
e75d1e4b61 Respect --test_output_files from DeepSpeech.py 2019-10-14 21:03:14 +02:00
Vinh Nguyen
909fa60601 adding automatic mixed precision training support 2019-10-14 12:34:10 +00:00
Alexandre Lissy
4f1f67f55d Use TFLite optimizations flag 
Fixes #2415
 2019-10-09 12:21:35 +02:00
lissyx
031479d88b
Merge pull request #2406 from lissyx/disable-cache-dataaug 
Disable cache when data augmentation is set
 2019-10-08 18:10:32 +02:00
Alexandre Lissy
c35068f880 Disable cache when data augmentation is set 
Fixes #2396
 2019-10-08 17:29:59 +02:00
Reuben Morais
5a287a65e5 Set forget_bias=0 in CPU graph for compatibility with CudnnRNN 2019-10-07 21:06:29 +02:00
Reuben Morais
4302a5f767 Make language model scoring optional in Python inference code 2019-09-30 11:43:00 +02:00
Reuben Morais
889f069b1c
Merge pull request #2352 from mozilla/data-augmentation-pr 
Data augmentation PR
 2019-09-09 21:29:02 +02:00
Bernardo Henz
49c6a9c973 adding 'train_phase' to create_dataset. Now we can augment only the training-set. 2019-09-09 12:07:51 +02:00
Reuben Morais
371e73eb69 Create checkpoint dir before writing flags file in it 2019-09-06 13:55:55 +02:00
Reuben Morais
ba2d29b36f Save flag values next to checkpoints 2019-09-06 11:35:10 +02:00
Reuben Morais
670e06365e Run tf_upgrade_v2 on our code 2019-08-28 17:53:24 +02:00
Reuben Morais
24bcdeb3d6 Switch from deprecated tfv1.app to absl-py 2019-08-28 10:55:33 +02:00
Reuben Morais
29836c9dcc
Merge pull request #2289 from mozilla/error-non-finite-loss 
Error message if a sample has non-finite loss
 2019-08-12 10:48:59 +02:00
Reuben Morais
86fff2f660
Merge pull request #2265 from mozilla/cudnnrnn_compatible 
Allow loading a CuDNN RNN checkpoint in a CPU-capable graph (Fixes #2264)
 2019-08-07 14:57:16 +02:00
Reuben Morais
248c01001e Error message if a sample has non-finite loss 2019-08-07 10:31:15 +02:00
Reuben Morais
c76070be19
Merge pull request #2282 from mozilla/dynamic-batch-size-in-train-val-graph 
Use dynamic batch size in train/val graph
 2019-08-07 10:03:53 +02:00
Reuben Morais
3636d9b481 Use dynamic batch size in train/val graph 
Avoid needing to use the same batch size for training and validation.
 2019-08-01 14:53:46 +02:00
Tilman Kamp
007e512c00 Fix #2180 - Added wav_filename to WER report 2019-07-23 16:36:10 +02:00