litagin02
|
867855ace9
|
Refactor, improve: use pathlib, allow audios other than wavs in resample
|
2024-03-15 11:10:23 +09:00 |
|
litagin02
|
5c1e791b9a
|
Add cleaning bat script
|
2024-03-15 10:30:22 +09:00 |
|
litagin02
|
73b01f7172
|
Add another wave dash for long symbol
|
2024-03-14 23:31:16 +09:00 |
|
litagin02
|
1b72023e02
|
Require torch>=2.1
|
2024-03-14 19:15:21 +09:00 |
|
litagin02
|
51e246658d
|
Change torch==2.1 to torch >=2.1
|
2024-03-14 18:30:21 +09:00 |
|
litagin02
|
2eb489f91e
|
docs
|
2024-03-14 18:29:06 +09:00 |
|
litagin02
|
6f42a1514f
|
Add ngram_size option
|
2024-03-14 17:42:29 +09:00 |
|
litagin02
|
8c95f0ff75
|
Fix hf whisper
|
2024-03-14 16:45:31 +09:00 |
|
litagin02
|
a5540bdb2e
|
Clean
|
2024-03-14 16:25:04 +09:00 |
|
litagin02
|
2996a0b195
|
docs
|
2024-03-14 15:41:10 +09:00 |
|
litagin02
|
b65d1d4755
|
Improve log and webui
|
2024-03-14 15:19:09 +09:00 |
|
litagin02
|
b1972a3d3d
|
Feat: HF whisper for transcribing (faster than faster-whisper)
|
2024-03-14 15:10:08 +09:00 |
|
litagin02
|
4f60a3d5d5
|
Feat: multiprocessing of slicing for much faster slicing
|
2024-03-13 18:56:51 +09:00 |
|
litagin02
|
dd407d882d
|
Fix typing of arguments (Languages class)
|
2024-03-13 18:45:25 +09:00 |
|
litagin02
|
67f19ba627
|
Fix: disable multiprocessing for bert_gen due to pyopenjtalk_worker
|
2024-03-13 17:12:12 +09:00 |
|
litagin02
|
b71e304fe9
|
Change SileroVAD to my repo to disable onnx warning
|
2024-03-13 16:52:21 +09:00 |
|
litagin02
|
aff666db75
|
Update webui for time_suffix
|
2024-03-13 16:51:23 +09:00 |
|
litagin02
|
1ddea1aa6d
|
Merge branch 'master' into dev
|
2024-03-13 16:48:49 +09:00 |
|
litagin02
|
f4d34b392c
|
Merge branch 'master' into dev
|
2024-03-13 16:35:53 +09:00 |
|
litagin02
|
9292270873
|
Merge pull request #98 from aka7774/master
Add time_suffix option
|
2024-03-13 16:35:29 +09:00 |
|
litagin02
|
860957006b
|
Add typing and load bert models on the top
|
2024-03-13 16:27:06 +09:00 |
|
litagin02
|
3141a8e341
|
debug logging
|
2024-03-13 16:26:29 +09:00 |
|
aka7774
|
3d36778e06
|
Add time_suffix option
|
2024-03-13 16:07:09 +09:00 |
|
litagin02
|
7e3e6c55fa
|
docs
|
2024-03-13 15:46:02 +09:00 |
|
litagin02
|
b309c9d74a
|
Change from 音程 to 音高
|
2024-03-13 15:43:33 +09:00 |
|
litagin02
|
3c1269e520
|
Add library usage example notebook
|
2024-03-13 15:41:46 +09:00 |
|
litagin02
|
cd02238b28
|
Add POST method to server_fastapi.py
|
2024-03-13 15:28:23 +09:00 |
|
litagin02
|
55c3847176
|
Add POST method for /voice in server_fastapi.py
|
2024-03-13 15:26:30 +09:00 |
|
litagin02
|
c028840794
|
Merge pull request #93 from litagin02/dev-refactor
Dev refactor
|
2024-03-13 15:12:02 +09:00 |
|
litagin02
|
bfcf5091ed
|
Fix running server_editor.py in bat (and dev ver)
|
2024-03-13 15:07:01 +09:00 |
|
litagin02
|
bc89fde15e
|
Delete local_dir_use_symlinks=False
|
2024-03-13 14:55:42 +09:00 |
|
tsukumi
|
9e5222619f
|
Refactor: No preloading of BERT models to avoid unnecessary GPU VRAM consumption during training in the Web UI
Since the BERT features of the dataset are pre-extracted by bert_gen.py, there is no need to load the BERT model at training time.
|
2024-03-13 12:09:11 +09:00 |
|
litagin02
|
8baecb138a
|
Skip cuda test
|
2024-03-13 11:54:24 +09:00 |
|
litagin02
|
c37893a76b
|
Add version suffix dev0
|
2024-03-13 11:46:56 +09:00 |
|
litagin02
|
28fd715cb8
|
Restore gc empty_cache in train_ms_jp
|
2024-03-13 10:25:22 +09:00 |
|
litagin02
|
30d9f2e258
|
Merge pull request #95 from zunan-islands/dev-refactor
細かな修正 (#92 を参照)
|
2024-03-13 09:45:55 +09:00 |
|
tsukumi
|
fedd019c04
|
Revert "Refactor: No preloading of BERT models to avoid unnecessary GPU VRAM consumption during training in the Web UI"
This reverts commit e8a76e547b.
|
2024-03-12 19:01:45 +00:00 |
|
tsukumi
|
c4d6a8cdb7
|
Fix: large number of unnecessary files in built sdist
Include in sdist only the minimum required files for style-bert-vits2 as a library.
|
2024-03-12 18:45:54 +00:00 |
|
tsukumi
|
07d246b98b
|
Refactor: run "hatch run style:fmt"
|
2024-03-12 18:27:08 +00:00 |
|
tsukumi
|
e8a76e547b
|
Refactor: No preloading of BERT models to avoid unnecessary GPU VRAM consumption during training in the Web UI
Since the BERT features of the dataset are pre-extracted by bert_gen.py, there is no need to load the BERT model at training time.
|
2024-03-12 18:22:34 +00:00 |
|
tsukumi
|
483bc68d57
|
Refactor: unify invoke format of open() function
|
2024-03-12 18:00:51 +00:00 |
|
tsukumi
|
74af2c831c
|
Add: cache_dir and revision optional arguments to load_model() / load_tokenizer()
|
2024-03-12 17:45:10 +00:00 |
|
litagin02
|
69225a3e13
|
Style
|
2024-03-12 21:47:59 +09:00 |
|
litagin02
|
6e4dcf0232
|
Add Data, model_assets to toplevel gitignore
|
2024-03-12 21:38:27 +09:00 |
|
litagin02
|
bdf394d2c2
|
Fix: migration to new TTSModelHolder
|
2024-03-12 20:56:50 +09:00 |
|
litagin02
|
528d2cc4ba
|
Refactor: typing and pathlib
|
2024-03-12 20:05:32 +09:00 |
|
litagin02
|
f12f20f3da
|
Rollback of SAFE_STDOUT debug
|
2024-03-12 19:13:13 +09:00 |
|
litagin02
|
4acfc2977e
|
Ensure to write use_jp_extra option to config.json
|
2024-03-12 19:10:43 +09:00 |
|
litagin02
|
456fe00ae4
|
Delete old webui bat files
|
2024-03-12 17:51:09 +09:00 |
|
litagin02
|
77ee026ab6
|
Update .dockerignore
|
2024-03-12 17:47:00 +09:00 |
|