iclementine
|
4d3014f4d5
|
add new trainer
|
2021-04-03 16:19:46 +08:00 |
|
iclementine
|
27e0201d0d
|
format code for tacotron_vctk, add plot_waveform to display
|
2021-04-02 15:46:28 +08:00 |
|
iclementine
|
a3fae49022
|
merge refactor_tacotron
|
2021-04-02 11:48:16 +08:00 |
|
iclementine
|
274d8dce07
|
update experiment and display
|
2021-04-02 11:37:48 +08:00 |
|
iclementine
|
15b205d6e0
|
Merge branch 'develop' into baker
|
2021-04-02 11:23:21 +08:00 |
|
chenfeiyu
|
9babec0f98
|
fix text log extention name
|
2021-04-01 13:49:52 +08:00 |
|
chenfeiyu
|
752272de98
|
fix bugs
|
2021-04-01 13:15:06 +08:00 |
|
iclementine
|
9798d07337
|
fix visualizer
|
2021-03-31 19:32:23 +08:00 |
|
iclementine
|
4a039b6407
|
add vctk example for refactored tacotron
|
2021-03-31 17:34:19 +08:00 |
|
iclementine
|
7cc3e8c340
|
add a simple strategy to support multispeaker for tacotron.
|
2021-03-31 15:23:41 +08:00 |
|
iclementine
|
2dd393349f
|
Merge branch 'develop' into refactor_tacotron
|
2021-03-30 16:01:22 +08:00 |
|
iclementine
|
e3f7bb5a51
|
simplify visualization code
|
2021-03-30 15:56:14 +08:00 |
|
chenfeiyu
|
0fdb86834b
|
Merge branch 'develop' into baker
|
2021-03-30 14:39:11 +08:00 |
|
chenfeiyu
|
b5dd0cc197
|
fix speaker encoder and add support for 2 more datasets
|
2021-03-30 14:38:44 +08:00 |
|
iclementine
|
4757f08550
|
Merge branch 'develop' into baker
|
2021-03-29 11:17:51 +08:00 |
|
iclementine
|
59ed247840
|
fix lstm speaker encoder
|
2021-03-29 11:17:23 +08:00 |
|
iclementine
|
ab85d5ca13
|
Merge branch 'develop' into baker
|
2021-03-29 11:13:57 +08:00 |
|
iclementine
|
5443e23fb7
|
fix lstm speaker encoder
|
2021-03-29 11:12:02 +08:00 |
|
iclementine
|
a9a78742fa
|
Merge branch 'develop' into baker
|
2021-03-29 10:42:17 +08:00 |
|
iclementine
|
2475da3322
|
add ge2e
|
2021-03-27 17:39:37 +08:00 |
|
chenfeiyu
|
a005cc88a3
|
WIP: baker
|
2021-03-27 12:43:03 +08:00 |
|
iclementine
|
2b62fbb614
|
1. change the default min value of LogMagnitude to 1e-5;
2. remove stop logit prediction from tacotron2 model.
|
2021-03-23 10:44:22 +08:00 |
|
iclementine
|
da63cfa42e
|
add an embedding layer.
|
2021-03-22 21:39:22 +08:00 |
|
iclementine
|
f9d6160916
|
add an option to normalize volume when loading audio.
|
2021-03-22 21:38:28 +08:00 |
|
iclementine
|
086fbf8e35
|
refactoring code
|
2021-03-22 21:23:46 +08:00 |
|
chenfeiyu
|
3c60fec900
|
remove bn in postnet
|
2021-02-27 03:26:41 +08:00 |
|
chenfeiyu
|
929165b64a
|
1. remove space from numericalized representation;
2. fix decoder paddign mask's unsqueeze dim.
|
2021-02-27 02:59:38 +08:00 |
|
chenfeiyu
|
ae9e218073
|
use emb add in tacotron2
|
2021-02-26 18:08:26 +08:00 |
|
chenfeiyu
|
40237c40b0
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Parakeet into baker
|
2021-02-26 11:07:03 +08:00 |
|
chenfeiyu
|
6f1c534557
|
use exist_ok to ensure no conflict
|
2021-02-23 10:50:57 +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 |
|
chenfeiyu
|
6a92fde9b2
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Parakeet into baker
|
2021-02-18 19:58:27 +08:00 |
Hui Zhang
|
c955c4192b
|
fix tensorboard error https://github.com/pytorch/fairseq/issues/2357
|
2021-02-18 19:53:58 +08:00 |
|
chenfeiyu
|
25bd8987a6
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Parakeet into baker
|
2021-02-18 19:51:56 +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 |
|
chenfeiyu
|
239703be8b
|
hacky thing, add tone support for acoustic model
|
2021-02-10 22:58:08 +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 |
|
iclementine
|
c321fcd098
|
polish documentation
|
2021-01-13 14:58:26 +08:00 |
|
iclementine
|
641be1bc92
|
Merge branch 'develop' of github.com:iclementine/Parakeet into doc
|
2021-01-13 11:09:05 +08:00 |
|
iclementine
|
c2a279c433
|
add documentation sections
|
2021-01-13 11:06:15 +08:00 |
lfchener
|
b0ba6e7bf9
|
fix an encoding problem in windows
|
2021-01-08 02:47:43 +00:00 |
|
lfchener
|
80bf04b710
|
fix the behavior of dropout in eval of tacotron2
|
2020-12-28 08:28:55 +00:00 |
|
chenfeiyu
|
2421a936ed
|
fix positional encoding naming conflict
|
2020-12-21 17:41:18 +08:00 |
iclementine
|
bb64e4659a
|
discard opencc untill we find an easy solution to install it on windows
|
2020-12-20 13:46:45 +08:00 |
|
iclementine
|
e03e96d9e4
|
format all the code with yapf
|
2020-12-20 13:15:07 +08:00 |
|
iclementine
|
f31643b33c
|
1. fix typos;
2. add tensorboardX into install requirements.
|
2020-12-19 20:08:25 +08:00 |
|
iclementine
|
18709adce8
|
update setup.py and version str
|
2020-12-19 18:55:42 +08:00 |
|
iclementine
|
b6efb43990
|
update docstring for waveflow
|
2020-12-19 18:33:07 +08:00 |
|
iclementine
|
f2a35a17d4
|
import normalizer into frontend
|
2020-12-19 16:20:41 +08:00 |