Commit Graph

  • 99e819e8ea Add ImpulseResponseAugmentor and augmentation.config file. Xinghai Sun 2017-07-04 18:51:11 +0800
  • ad82c87712 Add NoisePerturbAugmentor and CHiME3 data preparation. Xinghai Sun 2017-06-29 12:27:54 +0800
  • 961f6a2963 Accelerate mfcc computation for DS2. Xinghai Sun 2017-08-08 15:46:03 +0800
  • 6df0f9bc44 Reset default multi-thread/process number to half of cpu count() for speedup. Xinghai Sun 2017-08-08 12:15:18 +0800
  • f4375ef125 Update README.md with code reviews for DS2. Xinghai Sun 2017-08-07 19:41:59 +0800
  • 94db28e088 Add function doc for infer_batch_loss() function in model.py for DS2. Xinghai Sun 2017-08-03 21:06:27 +0800
  • c0b3281e58 Remove pynput and pyaudio packages from requriements.txt and add installation tips to README.md. Xinghai Sun 2017-08-03 19:00:17 +0800
  • 37b236869f Delete swig_decoder.py Yibing Liu 2017-08-03 14:46:18 +0800
  • b57d244363 Add ASR demo usage to README.md for DS2. Xinghai Sun 2017-08-03 12:26:33 +0800
  • a4c2dd7de2 Add function docs and comments to demo_server.py and demo_client.py. Xinghai Sun 2017-08-03 12:08:11 +0800
  • cb9370f308 Add warming-up to demo_server.py for DS2 and clean codes. Xinghai Sun 2017-08-03 12:00:24 +0800
  • 6bc445f235 refine the interface of decoders in swig Yibing Liu 2017-08-03 11:58:09 +0800
  • 0ebf36b98f Add a realtime ASR demo for users to test their own voice with mic. Xinghai Sun 2017-08-02 21:45:08 +0800
  • a3807d9cb5 Merge pull request #183 from xinghai-sun/refine_decoder2 Xinghai Sun 2017-08-03 01:57:57 +0800
  • 526e18b119 Add function docs for layer.py and model.py and update other details. Xinghai Sun 2017-08-02 23:50:46 +0800
  • 175457bf64 Merge pull request #185 from kuke/add_cudnn_info Yibing Liu 2017-08-02 16:03:21 +0800
  • a48469b9b6 add the requirement for cuDNN version in README Yibing Liu 2017-08-02 15:24:29 +0800
  • 5e20dfd4fb change the wget method in run.sh of deep_speech2 Luo Tao 2017-08-01 17:30:30 +0800
  • 8122dd9c29 Simplify train.py, evaluate.py, infer.py and tune.py by adding DeepSpeech2Model class. Xinghai Sun 2017-08-01 16:21:46 +0800
  • e2f954f5e2 make kenlm install more robust Luo Tao 2017-08-01 15:27:41 +0800
  • 92eacf548b Update default config params and result display for evaluator.py and infer.py for DS2. Xinghai Sun 2017-07-31 21:57:07 +0800
  • a840f85423 change probs' computation into log scale & add best path decoder Yibing Liu 2017-07-27 10:02:54 +0800
  • de212572ed update unittest with comments Luo Tao 2017-07-28 18:10:18 +0800
  • b72b70e54c add soundfile read/write unitest Luo Tao 2017-07-28 16:19:44 +0800
  • 7d7984b8fb add test for libsndfile installation Luo Tao 2017-07-27 15:40:08 +0800
  • ef5f0436f2 Clean warning logs in cloud/README.md wanghaoshuang 2017-07-27 14:13:14 +0800
  • 9fa9a352ac Refine submitting scripts for deepspeech2 on paddle cloud. wanghaoshuang 2017-07-27 13:53:37 +0800
  • bbe47a4318 Merge branch 'develop' of https://github.com/PaddlePaddle/models into ds2_pcloud wanghaoshuang 2017-07-26 20:02:47 +0800
  • c7676286ab install libsndfile from /usr to thirdparty Luo Tao 2017-07-26 16:25:11 +0800
  • cb0680e8c4 follow comments to modify README.md Yibing Liu 2017-07-20 11:47:46 +0800
  • 724ef18596 update several scripts to support mfcc Yibing Liu 2017-07-19 22:40:01 +0800
  • ee5abbe37d add mfcc feature for DS2 Yibing Liu 2017-07-18 10:14:37 +0800
  • ccea7c0150 enable loading language model in multiple format Yibing Liu 2017-07-10 11:34:47 +0800
  • 8ce9546710 modify README.md Yibing Liu 2017-07-11 14:56:37 +0800
  • d15c48d616 upload the language model Yibing Liu 2017-07-11 13:32:35 +0800
  • 5bfa066920 add setup and README for deployment Yibing Liu 2017-07-06 12:18:09 +0800
  • 94a6811660 code cleanup for the deployment decoder Yibing Liu 2017-07-06 11:25:05 +0800
  • 2e76f82cf7 Merge branch 'develop' into ctc_decoder_deploy Yibing Liu 2017-07-05 15:06:49 +0800
  • 0cadc56a84 follow comments in code format Yibing Liu 2017-07-05 11:05:26 +0800
  • 37e98df74d enable resetting params in scorer Yibing Liu 2017-07-04 19:48:56 +0800
  • d9d9514269 fix bugs Yibing Liu 2017-07-04 19:15:34 +0800
  • 3c77d369ca Make ds2 run on paddle cloud 1. Refine data_utils/data.py to read bytes from tar file 2. Add scripts to submit paddle cloud job for ds2 trainning wanghaoshuang 2017-07-03 15:14:23 +0800
  • fd102c2110 Merge branch 'develop' into ctc_decoder_deploy Yibing Liu 2017-06-29 15:56:25 +0800
  • 7c7e17e249 add deploy.py Yibing Liu 2017-06-29 11:19:39 +0800
  • 2f15a78707 add initial files for deployment Yibing Liu 2017-06-29 10:05:02 +0800
  • 90d83bf739 resolve conflicts in requirements.txt Yibing Liu 2017-06-27 20:27:13 +0800
  • aeccd9851b append README.md Yibing Liu 2017-06-27 20:22:47 +0800
  • db37c34919 modify some detail of augmentor xushaoyong 2017-06-27 18:48:49 +0800
  • 8ba98cb518 fix decoders' unittest Yibing Liu 2017-06-27 18:35:49 +0800
  • e1e2914ec9 remove augmentor unittest xushaoyong 2017-06-27 18:06:16 +0800
  • 7f2e7e1c2c resolve conflict with setup.sh xushaoyong 2017-06-27 17:48:25 +0800
  • 26510f74a6 refine ctc_beam_search_decoder Yibing Liu 2017-06-27 17:42:44 +0800
  • 286228c219 Merge pull request #127 from pkuyym/fix-soundfile xushaoyong 2017-06-27 17:25:17 +0800
  • 29f6ae0807 modify audio resample function xushaoyong 2017-06-27 17:16:35 +0800
  • c5995bb7e1 add 3 augmentor and unittest xushaoyong 2017-06-27 17:05:32 +0800
  • 237df23e7e Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev Yibing Liu 2017-06-26 22:01:59 +0800
  • 0dadd14600 Patch for adding missing shift_perturb.py in last commmit (pull request #114). Xinghai Sun 2017-06-26 19:19:16 +0800
  • 7324d41e4a Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev Yibing Liu 2017-06-26 17:48:59 +0800
  • cdd52ac270 Fix a missing abs bug for DS2 AudioSegment. Xinghai Sun 2017-06-26 14:17:22 +0800
  • 6d6cdf4057 Refine SoundFile installation process. 1. Install libsndfile first. 2. Install SoundFile using pip. yangyaming 2017-06-26 13:04:36 +0800
  • 13f708739b Improve audio featurizer and add shift augmentor. Xinghai Sun 2017-06-21 20:52:30 +0800
  • 8033845615 add unit test for decoders Yibing Liu 2017-06-21 14:52:06 +0800
  • b340d4ed2f modify setup.sh to delete the install of libsamplerate chrisxu2016 2017-06-21 12:18:33 +0800
  • d6a852a304 modify setup.sh to delete the install of libsamplerate chrisxu2016 2017-06-21 12:11:43 +0800
  • 2450591a44 add 3 augmentor class and change resample module chrisxu2016 2017-06-21 11:47:15 +0800
  • 5398360e5f Add 3 augmentor classes and related unittests chrisxu2016 2017-06-20 18:50:13 +0800
  • df77c6d5db Add 3 augmentor classes and related unittests chrisxu2016 2017-06-20 18:39:48 +0800
  • f20c6d643f Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev Yibing Liu 2017-06-20 18:21:11 +0800
  • d64f470078 add augmentor class chrisxu2016 2017-06-20 18:19:43 +0800
  • d104eccf67 Update the default num_threads for DS2 data generator. Xinghai Sun 2017-06-20 18:13:46 +0800
  • 1d8cc4a5a9 Add multi-threading support for DS2 data generator. Xinghai Sun 2017-06-20 17:06:53 +0800
  • 71283d619d add augmentor class chrisxu2016 2017-06-20 16:33:28 +0800
  • 115a06bb37 add augmentor class chrisxu2016 2017-06-20 16:24:03 +0800
  • 0729abe02e tiny adjust Yibing Liu 2017-06-20 12:14:24 +0800
  • d8089315f4 Merge branch 'develop' of https://github.com/PaddlePaddle/models into ctc_decoder_dev Yibing Liu 2017-06-20 12:03:30 +0800
  • a5dcd23bf2 Follow comments. yangyaming 2017-06-19 23:46:44 +0800
  • d35b747175 Fix ci following: 1. Unify the dependency installation process in setup.sh. 2. Change the version of package scipy from 0.13.0b1 to 0.13.1 yangyaming 2017-06-19 23:24:58 +0800
  • b2cf96cc46 adapt to the new data provider Yibing Liu 2017-06-19 19:00:08 +0800
  • d1f5123bcf Merge pull request #99 from PaddlePaddle/new_augmentation_2 Xinghai Sun 2017-06-19 15:30:04 +0800
  • def66a3223 Follow comments. yangyaming 2017-06-19 11:31:34 +0800
  • ddb2bdc190 add audio file chrisxu2016 2017-06-19 00:08:05 +0800
  • 25ce7ebe7b add audio file chrisxu2016 2017-06-18 18:22:48 +0800
  • 36743d3689 add scoring last word in beam search Yibing Liu 2017-06-18 18:11:01 +0800
  • 21161b0165 add audio file chrisxu2016 2017-06-18 17:11:58 +0800
  • 107f8b89ae add audio augmentation chrisxu2016 2017-06-18 16:47:09 +0800
  • ff01d048d3 final refining on old data provider: enable pruning & add evaluation & code cleanup Yibing Liu 2017-06-18 16:36:52 +0800
  • b8341da63d add audio augmentation chrisxu2016 2017-06-18 16:23:30 +0800
  • 26eb54eb37 Follow comments. yangyaming 2017-06-18 14:31:57 +0800
  • 86eeb52c06 Merge branch 'develop' of https://github.com/PaddlePaddle/models into fix-81 yangyaming 2017-06-18 13:55:37 +0800
  • 5ca270d30a add audio file chrisxu2016 2017-06-17 09:03:18 +0800
  • d1ee10be10 modify audio and speech chrisxu2016 2017-06-16 18:29:56 +0800
  • d66d740ea0 add audio part chrisxu2016 2017-06-16 15:17:43 +0800
  • ed5f04afb8 Add shuffle type of instance_shuffle and batch_shuffle_clipped. Xinghai Sun 2017-06-15 17:05:00 +0800
  • 65e34c535b add augmentation chrisxu2016 2017-06-15 03:08:30 +0800
  • 04a225ae4f Enable min_batch_num in train.py and update train info print. Xinghai Sun 2017-06-14 18:14:50 +0800
  • f85f8558cf Add unittest. yangyaming 2017-06-14 15:00:10 +0800
  • 397b2fc288 Merge branch 'develop' of https://github.com/PaddlePaddle/models into fix-81 yangyaming 2017-06-14 14:46:32 +0800
  • 1cef98f210 Update README.md for DS2. Xinghai Sun 2017-06-13 23:33:38 +0800
  • b07ee84a1d Add function, class and module docs for data parts in DS2. Xinghai Sun 2017-06-13 23:16:07 +0800
  • a633eb9cc6 change two arguments Yibing Liu 2017-06-13 15:28:43 +0800