Commit Graph

  • 012f159ab5 Fix generating style bug litagin02 2024-05-26 11:07:43 +09:00
  • e52453040c Improve: transc order by path, train.list preserves order litagin02 2024-05-26 10:47:14 +09:00
  • d8ca829e28 Docs, if only one subdir skip generating style litagin02 2024-05-26 10:24:52 +09:00
  • ce26a0384d Improve: save intermediate trans result for HF whisper litagin02 2024-05-26 10:18:09 +09:00
  • f1e022a08c Fix tb logging litagin02 2024-05-26 08:27:39 +09:00
  • 5fa6210176 Batch sampler for backward compatibility, reduce tb log litagin02 2024-05-26 08:23:38 +09:00
  • a92b0cabff Improve: slice.py preserves subdir structure litagin02 2024-05-26 08:22:39 +09:00
  • 7880659d4a Docs litagin02 2024-05-26 08:00:32 +09:00
  • a4f28e4f6f Improve download litagin02 2024-05-25 19:28:22 +09:00
  • 5af4e2c6ac Feat: download model from hf litagin02 2024-05-25 19:05:06 +09:00
  • 2274087da4 Fmt only (maybe) litagin02 2024-05-25 18:15:36 +09:00
  • acf93b80ce Feat: make style vec by subdirs (for webui) litagin02 2024-05-25 18:13:34 +09:00
  • 175aa62a26 Feat: make style vectors w.r.t. subdirs structure litagin02 2024-05-25 17:14:28 +09:00
  • 0c87c6ffd4 Clean, and fix tensorboard protobuf bug litagin02 2024-05-25 15:58:19 +09:00
  • 8976a3ed2f Refactor: Use PathConfig and pathlib instead of paths.yml loading litagin02 2024-05-25 10:25:22 +09:00
  • 2771fbd209 Fix gradio Path serialization error litagin02 2024-05-22 07:31:40 +09:00
  • 053a6bf785 Fix gradio Path serialize error litagin02 2024-05-19 18:31:45 +09:00
  • 0a3b57528b Fix gradio import type annotation bug (temp solution) litagin02 2024-05-15 20:39:22 +09:00
  • ac90941eaa Remove gradio version restriction litagin02 2024-05-15 19:23:38 +09:00
  • 837c4397db Merge pull request #118 from tsukumijima/dev litagin02 2024-05-15 07:37:46 +09:00
  • ae7c4ea72b Merge pull request #124 from gordon0414/eng_g2p_fix litagin02 2024-05-15 07:01:02 +09:00
  • 44693e7d6b Refactor: improve regular expression performance during NLP Improve performance by pre-compiling regular expressions that are executed many times. tsukumi 2024-05-12 00:26:46 +09:00
  • 7de367b76f moved tokenizer loading inside of g2p function for code integrity. gordon0414 2024-05-10 11:55:38 +09:00
  • 3018309199 Optimize g2p function for improved speed and efficiency gordon0414 2024-05-09 16:12:15 +09:00
  • 2d6baa3928 Fix: changed the substitute phoneme from「ん」to 「'」 when the pronunciation can't be obtained tsukumi 2024-05-06 22:17:34 +09:00
  • 87144dd321 Improve: automatically generate configs/paths.yml by copying it from configs/default_paths.yml when running initialize.py If configs/paths.yml itself is included in version control, differences will occur when it is changed in each environment, which is troublesome. tsukumi 2024-05-06 20:56:07 +09:00
  • b230832528 Add: allow hyper-parameters and style vectors to be specified directly tsukumi 2024-05-06 20:43:22 +09:00
  • 0d9034b8bd Fix text encoding in Server.bat tsukumi 2024-05-01 11:21:01 +09:00
  • 3730eb50b7 Merge branch 'litagin02:dev' into dev tsukumi 2024-05-01 10:56:42 +09:00
  • e721c09b87 Fix filename for transcribe to use relative path instead of name litagin02 2024-04-27 12:59:47 +09:00
  • 9df344b995 Fix nested audio files bug litagin02 2024-04-27 12:50:00 +09:00
  • aa0f9308b1 Refactor: remove dependencies on libraries not required for style_bert_vits2 as a library librosa: Originally not used under style_bert_vits2/. pyannote.audio: Required only when using the "infer style vector from audio" function during inference, but it is currently almost unused. scipy: Because it was used in only one utility function that was not needed outside of training. tsukumi 2024-04-25 00:31:14 +00:00
  • 32145e18a3 Merge pull request #111 from hopto-dot/master litagin02 2024-04-23 19:21:15 +09:00
  • 28f254f449 Improve: remove Gradio dependency from style_bert_vits2 as a library tsukumi 2024-04-23 05:45:11 +00:00
  • f00344944d Merge pull request #117 from aka7774/master litagin02 2024-04-22 21:39:57 +09:00
  • a07609cb83 Merge pull request #115 from Naozumi520/patch-1 litagin02 2024-04-22 21:39:32 +09:00
  • 9d15d5763d Refactor: run "hatch run style:fmt" tsukumi 2024-04-19 04:17:27 +00:00
  • 797c354b1b Add: the ability to specify a phoneme sequence representing the reading of the text to be read during speech synthesis With this feature, it is possible to specify individual readings for homonyms such as "明日 (ashita/asu)" while maintaining the Kanji representation of the read text. tsukumi 2024-04-19 04:09:21 +00:00
  • 590937d71b Add files via upload Aka Diffusion 2024-04-16 14:52:03 +09:00
  • aba7814580 Removed a redundancy bracket in readme.md Naozumi 2024-04-12 13:41:57 +08:00
  • 556f8e49f3 fixed link to Style-Bert-VITS2_en.md GC 2024-04-04 12:53:28 +01:00
  • f94896420c Added article translation to en readme GC 2024-04-04 12:51:55 +01:00
  • 59063cde70 Added english translation of https://zenn.dev/litagin/articles/034819a5256ff4 GC 2024-04-04 12:47:45 +01:00
  • 18876a077a Delete Style-Bert-VITS2 JP-Extra (EN).md JawGBoi 2024-04-04 12:45:09 +01:00
  • 962ea6aee2 Added translation of article JawGBoi 2024-04-04 12:43:29 +01:00
  • 7a34fd24c1 Fix gradio ver for json serialize error litagin02 2024-03-29 21:51:04 +09:00
  • 6d5b854cf2 Fix: invalid path for generating neutral style litagin02 2024-03-22 15:54:22 +09:00
  • 4afe92d75f Fix: gen style in beginning of train for nested folders litagin02 2024-03-22 08:27:43 +09:00
  • d7a3e5696c docs litagin02 2024-03-19 20:16:07 +09:00
  • 70d252bcc7 Feat: preserve relative path when resample litagin02 2024-03-19 20:14:19 +09:00
  • 7c484c7835 Powershell version litagin02 2024-03-19 08:51:41 +09:00
  • 8c9eb74335 Fix colab litagin02 2024-03-18 09:02:50 +09:00
  • 8d6eec9d1b docs litagin02 2024-03-16 20:25:15 +09:00
  • defb5253aa Fix errorcheck litagin02 2024-03-16 18:41:29 +09:00
  • 251b075942 Remove debug pause litagin02 2024-03-16 18:27:28 +09:00
  • baeecbc2b9 Clean litagin02 2024-03-16 18:25:56 +09:00
  • 94e29d1337 Remove debug pause litagin02 2024-03-16 18:25:33 +09:00
  • 76be727b2e Bump ver litagin02 2024-03-16 18:22:35 +09:00
  • b363a8866e Remove debugging pause litagin02 2024-03-16 18:03:04 +09:00
  • 242e7ee2f0 Fix activating venv litagin02 2024-03-16 17:26:05 +09:00
  • f2cc879603 Fix: eneabledelayedexpansion for error code litagin02 2024-03-16 16:49:22 +09:00
  • 5eef8843cc Improve bat, replace MinGit with PortableGit litagin02 2024-03-16 15:25:19 +09:00
  • 8a8cd1dcaf Feat: using MinGit for bat install and update litagin02 2024-03-16 11:56:48 +09:00
  • 386d122cc7 Change webui/ to gradio_tabs/ to avoid conflict litagin02 2024-03-15 22:41:19 +09:00
  • 9986ce3ca4 Merge branch 'dev' litagin02 2024-03-15 21:50:20 +09:00
  • 5bafd7ffe9 docs litagin02 2024-03-15 21:50:04 +09:00
  • c8f4b82157 Merge pull request #97 from litagin02/dev litagin02 2024-03-15 21:44:39 +09:00
  • ab4020545e Bump torch ver and docs litagin02 2024-03-15 21:43:57 +09:00
  • 14c6805337 Update hatch setting and fmt litagin02 2024-03-15 21:36:53 +09:00
  • 2ee4f4ac86 Change dev to master in bat litagin02 2024-03-15 21:27:22 +09:00
  • 1a8a7edb9c Docs and bump default pytorch version to 2.2.1 latest litagin02 2024-03-15 21:25:24 +09:00
  • e42991f7e3 Remove loading bert since necessary model is loaded automatically litagin02 2024-03-15 21:10:24 +09:00
  • f796af5fbb Remove loading bert since necesary model is loaded automatically litagin02 2024-03-15 21:07:56 +09:00
  • 5100b4a57e docs litagin02 2024-03-15 20:30:22 +09:00
  • 3efcb98858 docs litagin02 2024-03-15 20:16:58 +09:00
  • d808cd635d Delete num_process option for slice for simplicity litagin02 2024-03-15 19:58:16 +09:00
  • d808793960 Change default HF whisper batch size to 16 litagin02 2024-03-15 19:56:55 +09:00
  • b02b882c04 Feat: support various audio format in slice.py litagin02 2024-03-15 12:53:24 +09:00
  • 432edcc52b Bump ver, rename configs_jp_extra to config_jp_extra for consistency litagin02 2024-03-15 12:34:54 +09:00
  • 3d8b60c03c Refactor litagin02 2024-03-15 12:27:46 +09:00
  • 867855ace9 Refactor, improve: use pathlib, allow audios other than wavs in resample litagin02 2024-03-15 11:10:23 +09:00
  • 5c1e791b9a Add cleaning bat script litagin02 2024-03-15 10:30:22 +09:00
  • 73b01f7172 Add another wave dash for long symbol litagin02 2024-03-14 23:31:16 +09:00
  • 1b72023e02 Require torch>=2.1 litagin02 2024-03-14 19:15:21 +09:00
  • 1d9b91fd2a Update torch req to torch>=2.1 from torch==2.1.x litagin02 2024-03-14 19:02:55 +09:00
  • 51e246658d Change torch==2.1 to torch >=2.1 litagin02 2024-03-14 18:30:21 +09:00
  • 2eb489f91e docs litagin02 2024-03-14 18:29:06 +09:00
  • 6f42a1514f Add ngram_size option litagin02 2024-03-14 17:42:29 +09:00
  • 8c95f0ff75 Fix hf whisper litagin02 2024-03-14 16:45:31 +09:00
  • a5540bdb2e Clean litagin02 2024-03-14 16:25:04 +09:00
  • 2996a0b195 docs litagin02 2024-03-14 15:41:10 +09:00
  • b65d1d4755 Improve log and webui litagin02 2024-03-14 15:19:09 +09:00
  • b1972a3d3d Feat: HF whisper for transcribing (faster than faster-whisper) litagin02 2024-03-14 15:10:08 +09:00
  • 4f60a3d5d5 Feat: multiprocessing of slicing for much faster slicing litagin02 2024-03-13 18:56:51 +09:00
  • dd407d882d Fix typing of arguments (Languages class) litagin02 2024-03-13 18:45:25 +09:00
  • 67f19ba627 Fix: disable multiprocessing for bert_gen due to pyopenjtalk_worker litagin02 2024-03-13 17:12:12 +09:00
  • b71e304fe9 Change SileroVAD to my repo to disable onnx warning litagin02 2024-03-13 16:52:21 +09:00
  • aff666db75 Update webui for time_suffix litagin02 2024-03-13 16:51:23 +09:00
  • 1ddea1aa6d Merge branch 'master' into dev litagin02 2024-03-13 16:48:49 +09:00
  • f4d34b392c Merge branch 'master' into dev litagin02 2024-03-13 16:35:53 +09:00