Commit Graph

507 Commits

Author SHA1 Message Date
chenfeiyu 58a988c789 add code to compute statistics 2021-06-16 16:18:13 +00:00
chenfeiyu 27d3585606 add some profiling to unittesting 2021-06-16 14:42:11 +00:00
chenfeiyu 683cc1d30f add unit testings back 2021-06-16 13:49:18 +00:00
chenfeiyu 042e02d242 use paddle's profiler 2021-06-16 13:43:13 +00:00
chenfeiyu 8dbcc9bccb add profiling 2021-06-16 09:40:47 +00:00
chenfeiyu bbbe5a8b50 add profiling 2021-06-16 14:47:30 +08:00
chenfeiyu 95f64c4f02 WIP: add some trainig info 2021-06-14 17:21:45 +08:00
chenfeiyu b0983e4d76 WIP: pwg training works 2021-06-14 17:05:37 +08:00
chenfeiyu 54c7905f40 WIP: training setup done 2021-06-13 17:24:15 +08:00
chenfeiyu 0067851950 fix imports 2021-06-11 05:26:22 +08:00
chenfeiyu 66062d29e5 WIP: add sample code for parallel wavegan 2021-06-11 02:43:10 +08:00
chenfeiyu 258083aea9 fix file names 2021-06-11 01:01:43 +08:00
chenfeiyu f0a5ac8c5a move docstring to class 2021-06-11 00:59:07 +08:00
chenfeiyu 13ab0bd608 remove task-specific Datasets and add a general purpose DataTable 2021-06-10 22:58:53 +08:00
chenfeiyu 3bf2e71734 checkpoint utility: add support for str in addition to Path object 2021-06-10 20:34:59 +08:00
chenfeiyu 60c16dcfb7 add AudioMelDataset for training vocoders. 2021-06-10 20:16:14 +08:00
chenfeiyu 988d6d3268 add h5py utility to interact with numpy 2021-06-10 18:57:46 +08:00
chenfeiyu 8e31783b51 Merge branch 'develop' of https://github.com/PaddlePaddle/Parakeet into pwg 2021-06-10 16:26:34 +08:00
chenfeiyu 6a8b3f92df add AudioDataset and MelDataset 2021-06-10 16:26:09 +08:00
Feiyu Chan a0763986e2
add kbest (#114)
* add kbest

* fix typos

* remove unnecessary imports

* fix del fn

* fix typos

* add k-latest
2021-06-10 10:36:14 +08:00
chenfeiyu 3c964fde54 add parallel wavegan model 2021-06-10 04:08:05 +08:00
chenfeiyu 759999c738 STFT and MelScale: register filters as buffer. 2021-06-10 04:06:06 +08:00
chenfeiyu c306f5c2b3 add k-latest 2021-06-10 03:39:54 +08:00
chenfeiyu dc9040dd4d fix typos 2021-06-10 02:45:45 +08:00
chenfeiyu 0114a808a2 fix del fn 2021-06-10 00:40:54 +08:00
chenfeiyu 13323bdf6a remove unnecessary imports 2021-06-10 00:32:49 +08:00
chenfeiyu 37a66f1506 fix typos 2021-06-10 00:29:25 +08:00
chenfeiyu b571b506c3 add kbest 2021-06-10 00:25:00 +08:00
Feiyu Chan 87a0cec0c7
Merge pull request #111 from iclementine/develop
fix a config key error
2021-05-18 18:59:53 +08:00
chenfeiyu 3aa6aed0fb fix a config key error 2021-05-18 18:57:13 +08:00
Feiyu Chan 4f288a6d4f
add ge2e and tacotron2_aishell3 example (#107)
* hacky thing, add tone support for acoustic model

* fix experiments for waveflow and wavenet, only write visual log in rank-0

* use emb add in tacotron2

* 1. remove space from numericalized representation;
2. fix decoder paddign mask's unsqueeze dim.

* remove bn in postnet

* refactoring code

* add an option to normalize volume when loading audio.

* add an embedding layer.

* 1. change the default min value of LogMagnitude to 1e-5;
2. remove stop logit prediction from tacotron2 model.

* WIP: baker

* add ge2e

* fix lstm speaker encoder

* fix lstm speaker encoder

* fix speaker encoder and add support for 2 more datasets

* simplify visualization code

* add a simple strategy to support multispeaker for tacotron.

* add vctk example for refactored tacotron

* fix indentation

* fix class name

* fix visualizer

* fix root path

* fix root path

* fix root path

* fix typos

* fix bugs

* fix text log extention name

* add example for baker and aishell3

* update experiment and display

* format code for tacotron_vctk, add plot_waveform to display

* add new trainer

* minor fix

* add global condition support for tacotron2

* add gst layer

* add 2 frontend

* fix fmax for example/waveflow

* update collate function, data loader not does not convert nested list into numpy array.

* WIP: add hifigan

* WIP:update hifigan

* change stft to use conv1d

* add audio datasets

* change batch_text_id, batch_spec, batch_wav to include valid lengths in the returned value

* change wavenet to use on-the-fly prepeocessing

* fix typos

* resolve conflict

* remove imports that are removed

* remove files not included in this release

* remove imports to deleted modules

* move tacotron2_msp

* clean code

* fix argument order

* fix argument name

* clean code for data processing

* WIP: add README

* add more details to thr README, fix some preprocess scripts

* add voice cloning notebook

* add an optional to alter the loss and model structure of tacotron2, add an alternative config

* add plot_multiple_attentions and update visualization code in transformer_tts

* format code

* remove tacotron2_msp

* update tacotron2 from_pretrained, update setup.py

* update tacotron2

* update tacotron_aishell3's README

* add images for exampels/tacotron2_aishell3's README

* update README for examples/ge2e

* add STFT back

* add extra_config keys into the default config of tacotron

* fix typos and docs

* update README and doc

* update docstrings for tacotron

* update doc

* update README

* add links to downlaod pretrained models

* refine READMEs and clean code

* add praatio into requirements for running the experiments

* format code with pre-commit

* simplify text processing code and update notebook
2021-05-13 17:49:50 +08:00
Hui Zhang 0aa7088d36
Merge pull request #97 from iclementine/develop
fix multiprocess training
2021-02-23 10:52:01 +08:00
chenfeiyu 6f1c534557 use exist_ok to ensure no conflict 2021-02-23 10:50:57 +08:00
chenfeiyu c6504ade5a Merge branch 'develop' of https://github.com/PaddlePaddle/Parakeet into develop 2021-02-23 10:40:26 +08:00
chenfeiyu e69ab88fe6 fix multiprocessing training: other processes have to wait untils the output directory in created. 2021-02-23 10:40:14 +08:00
Feiyu Chan dc055bde0a
Merge pull request #96 from iclementine/develop
fix experiments for waveflow and wavenet
2021-02-22 14:22:42 +08:00
chenfeiyu 7b0de356f9 fix experiments for waveflow and wavenet, only write visual log in rank-0 2021-02-21 17:34:11 +08:00
Hui Zhang 3ebe5ccb33
Merge pull request #95 from PaddlePaddle/fix
fix tensorboard error https://github.com/pytorch/fairseq/issues/2357
2021-02-18 19:58:21 +08:00
Hui Zhang c955c4192b fix tensorboard error https://github.com/pytorch/fairseq/issues/2357 2021-02-18 19:53:58 +08:00
Feiyu Chan a3de28cbe0
Merge pull request #94 from iclementine/develop
fix bugs with multiprocess training.
2021-02-18 19:48:56 +08:00
chenfeiyu 0af7402daa add rank_zero_only for ExperimentBase.save 2021-02-18 19:33:41 +08:00
chenfeiyu f423323bae fix bugs with multiprocess training. 2021-02-18 19:09:54 +08:00
Feiyu Chan 3f60b6e0a3
Merge pull request #92 from iclementine/develop
sevral fixes to transformer tts
2021-02-03 15:07:09 +08:00
chenfeiyu 30e3b9172f 1. fix imports for renamed functions in position encoding;
2. fix dimension in MLPPrenet in transformer_tts;
3. use dropout also in inference in MLPPrenet in transformer_tts.
2021-02-03 14:24:29 +08:00
Feiyu Chan a0ce65211c
Merge pull request #86 from iclementine/doc
update doc and README
2021-01-18 15:20:26 +08:00
iclementine c5acfbd8eb fix typos 2021-01-18 15:15:49 +08:00
iclementine 087d7bf16e remove dead links 2021-01-18 13:13:56 +08:00
iclementine cf892c5ed7 add paddlepaddle into requirements for readthedocs 2021-01-14 15:30:31 +08:00
iclementine 73374528d0 add tutorials into sdvanced 2021-01-14 15:12:36 +08:00
iclementine b017c73100 remove mocking of paddle, fix typos 2021-01-14 13:17:29 +08:00