Commit Graph

Select branches

Hide Pull Requests

master

patch

#1

#2

Mono Color

• db37175b87 fix patch tuna2134 2026-07-26 00:49:25 +09:00
• 0d6eebc3a8 fix tuna2134 2026-07-24 16:29:09 +09:00
• 8125666e22 split tuna2134 2026-07-24 16:24:50 +09:00
• 7e20916d1a merge master tuna2134 2026-07-24 16:06:49 +09:00
• 1785e81509 fix tuna2134 2026-07-23 15:59:17 +09:00
• a8b36ba33f fix tuna2134 2026-07-23 14:51:23 +09:00
• fd72a19384 bump tuna2134 2026-07-23 12:06:17 +09:00
• 9cfaa27f3f fix tuna2134 2026-07-23 11:51:07 +09:00
• 4f3bcfd252 fix tuna2134 2026-07-23 11:33:43 +09:00
• 7d535f3761 fix tuna2134 2026-07-20 21:29:02 +09:00
• b8ce11605c fix tuna2134 2026-07-20 21:25:49 +09:00
• 0c2be00f0a ok tuna2134 2026-07-20 21:08:44 +09:00
• 66de777e06 Zip file to latest file litagin02 2025-08-24 12:04:16 +09:00
• d8148f3090 Merge pull request #213 from litagin02/dev litagin02 2025-08-24 11:57:07 +09:00
• bfb2435b37 Merge branch 'master' into dev litagin02 2025-08-24 11:55:14 +09:00
• 4ea3cf12ef update readme litagin02 2025-08-24 11:48:21 +09:00
• e5f2dafbdd update litagin02 2025-08-24 11:46:53 +09:00
• 998217c010 fmt, credit litagin02 2025-08-24 11:39:39 +09:00
• c51c80fe1a Ver 2.7.0 maybe litagin02 2025-08-24 11:27:27 +09:00
• 4859d64e9f Change onnxsim to onnxsim-prebuilt litagin02 2025-08-24 09:40:38 +09:00
• c55bc61a45 Add python version file (for e.g. uv?) litagin02 2025-08-24 09:39:54 +09:00
• 4eab125451 Sort model files by desceinding mt time litagin02 2025-08-24 09:39:27 +09:00
• f8834b4c74 Merge pull request #193 from kale4eat/fix-pyopenjtalk-worker litagin02 2025-08-24 09:37:36 +09:00
• 1d7a7a0d48 Fix colab numpy version issue litagin02 2025-03-29 22:40:22 +09:00
• 404b2811de Merge pull request #194 from tsukumijima/master litagin02 2025-03-29 22:06:35 +09:00
• d7e2745559 Improve: Update license information tsukumi 2025-03-26 20:28:34 +09:00
• 1a44c9d437 Merge branch 'dev' of github.com:litagin02/Style-Bert-VITS2 tsukumi 2025-03-26 20:09:40 +09:00
• 2aa503233a Improve: Add option to generate AIVM/AIVMX files directly when running convert_onnx.py, add license information tsukumi 2025-03-26 20:01:17 +09:00
• 636f20fcfe Fix startup error by enabling SO_REUSEADDR option kale4eat 2024-12-24 21:36:44 +09:00
• 810ca43615 Improve: Disable enable_cpu_mem_arena for CPU inference only to prevent excessive memory consumption during the inference session of the BERT model tsukumi 2024-12-22 09:17:31 +09:00
• 3c218c7087 Improve: Variation of text-to-speech during testing tsukumi 2024-12-22 07:17:07 +09:00
• 405c5deddd Improve: Adjusted the default Execution Provider options tsukumi 2024-12-22 07:04:20 +09:00
• 08c439e88e Refactor: Use I/O Binding during BERT inference and always release memory after inference tsukumi 2024-12-22 06:49:43 +09:00
• 15af441cba Improve: Convert ONNX version of BERT models to FP16 tsukumi 2024-12-19 14:51:30 +09:00
• b843804a75 Fix: TTSModel.unload() did not work in PyTorch-independent environments tsukumi 2024-12-19 13:51:52 +09:00
• e1ce12a336 Fix: Use Fast Tokenizer instead of Slow Tokenizer, which makes validation very slow tsukumi 2024-12-19 13:41:09 +09:00
• ebd249bca8 Fix: spm.model is missing tsukumi 2024-12-19 12:05:33 +09:00
• 2833fb5eeb Fix: The conversion script for English BERT to Fast Tokenizer was incorrect, so tokenization was not performed correctly tsukumi 2024-12-19 11:10:33 +09:00
• c1fce3fec7 Improve: Make it possible to convert BERT language models to FP16 tsukumi 2024-12-19 07:16:59 +09:00
• e642a4cb69 Merge branch 'litagin02:master' into master tsukumi 2024-12-08 03:37:27 +09:00
• 3383458317 delete unrelated file litagin02 2024-11-20 18:58:15 +09:00
• 9740256c87 Add anime-whisper litagin02 2024-11-17 16:47:37 +09:00
• 8d476175a7 hathc run style:fmt litagin02 2024-11-17 12:44:27 +09:00
• a6bc65271c hatch fmt litagin02 2024-11-17 12:41:05 +09:00
• a6047012dd Merge pull request #178 from tsukumijima/master litagin02 2024-11-17 12:35:17 +09:00
• 065a7ffa0a Merge pull request #177 from aka7774/master litagin02 2024-11-17 12:35:02 +09:00
• 4a92a1abad Fix: Missing onnxsim dependency tsukumi 2024-11-13 10:50:44 +09:00
• f135a79316 Fix: Missing onnx dependency tsukumi 2024-11-13 10:43:58 +09:00
• 875186f0c7 Change faster-whisper as default litagin02 2024-11-11 10:25:30 +09:00
• 038d07f4e7 Bump ver only litagin02 2024-11-11 10:18:55 +09:00
• 2cb5b24244 Merge pull request #165 from tsukumijima/master litagin02 2024-11-11 09:44:32 +09:00
• 3155e2acc1 Update server_fastapi.py Aka Diffusion 2024-11-10 15:27:19 +09:00
• 689217e350 Refactor: TTSModel.convert_to_16_bit_wav() to @staticmethod tsukumi 2024-11-10 07:11:53 +09:00
• 20b0b0908b Fix: Ensure memory is cleared on unload tsukumi 2024-11-10 07:06:20 +09:00
• 9723f75f99 Fix: g2p process fails in PyTorch-independent environment tsukumi 2024-11-08 02:56:05 +09:00
• b0b12696f3 Add: Type Hint to null_model_params tsukumi 2024-10-25 22:37:09 +09:00
• 375eb7f3a9 Refactor: run "hatch run style:fmt" tsukumi 2024-10-25 22:18:57 +09:00
• d762bf4401 Merge branch 'dev' of github.com:litagin02/Style-Bert-VITS2 tsukumi 2024-10-25 22:16:19 +09:00
• d397a58874 Fix: errors during ONNX CPU inference tsukumi 2024-10-06 16:18:21 +09:00
• c4b6c169fc Fix: Inference failure with some ExecutionProviders tsukumi 2024-09-28 17:36:26 +09:00
• 7793e9da43 Remove: Unused import tsukumi 2024-09-28 13:50:10 +09:00
• d935d30e4c Add: test code for CoreMLExecutionProvider on a trial basis tsukumi 2024-09-23 21:07:27 +09:00
• b67e50fd8d Improve: DirectML inference performance tsukumi 2024-09-23 20:20:46 +09:00
• 08af692835 Fix: PyTorch inference test tsukumi 2024-09-23 15:20:52 +09:00
• e91b69862c Fix: ONNX inference with DirectML fails tsukumi 2024-09-23 15:06:19 +09:00
• 758696c59c Fix: Device ID specified in ExecutionProvider options is respected when moving input tensors tsukumi 2024-09-23 12:40:07 +09:00
• c792bf9a0d Improve: Make ONNX inference tests more rigorous tsukumi 2024-09-23 12:27:29 +09:00
• 2cdb5844cb Fix: ONNX inference fails in Windows environments tsukumi 2024-09-23 11:55:50 +09:00
• 53777e315e Fix: Log in real time during pytest execution tsukumi 2024-09-23 11:39:10 +09:00
• 5ca6e184d7 Fix: ONNX inference of BERT models fails in some Windows environments tsukumi 2024-09-23 11:33:17 +09:00
• f9607ad891 Fix: Omitted consideration of macOS support tsukumi 2024-09-23 11:10:41 +09:00
• f21f6435ad Improve: Add test code for ONNX inference with DmlExecutionProvider tsukumi 2024-09-23 10:54:29 +09:00
• eb5d245903 Add: Test for ONNX inference code (without PyTorch dependency) tsukumi 2024-09-23 10:04:02 +09:00
• 5a82105df3 Improve: Since the use cases for speech synthesis in English and Chinese are limited, only Japanese BERT models/tokenizers will be pre-loaded to save VRAM tsukumi 2024-09-23 08:58:57 +09:00
• faddb0a8a8 Fix: Don't list hidden files as models tsukumi 2024-09-23 08:22:20 +09:00
• 245bfe54ae Fix: In the “Merge” tab of the Web UI, re-create the TTSModelHolder using the passed TTSModelHolder instance variable so that ONNX models do not get mixed up tsukumi 2024-09-23 08:13:16 +09:00
• b7cd512978 Fix: During ONNX inference, the tensor of the return value of the tokenizer is now obtained in Numpy's NDArray format to eliminate the dependency on PyTorch tsukumi 2024-09-23 07:40:17 +09:00
• dcca37b9a8 Add: Support for speech synthesis in English and Chinese for ONNX inference in Non-JP-Extra models tsukumi 2024-09-23 07:14:44 +09:00
• fe253611d3 Fix: nltk to 3.8.1 or lower tsukumi 2024-09-23 06:22:47 +09:00
• db0acd7b5f Add: Support for ONNX inference and ONNX conversion of Non-JP-Extra models tsukumi 2024-09-23 05:59:12 +09:00
• cc1d6120fe Fix: Default value of False for compatibility with older models where the HyperParameters.data.use_jp_extra field does not exist tsukumi 2024-09-23 05:26:19 +09:00
• 0e0b4bea31 Fix: Disable optimization on InferenceSession initialization to speed up loading of ONNX models tsukumi 2024-09-22 23:08:00 +09:00
• 16115a366b Improve: .safetensors models under a directory specified with --model can be automatically converted to ONNX tsukumi 2024-09-22 16:27:22 +09:00
• 291e635e9e Improve: ONNX conversion logic improved, and support for noise_scale(_w) model argument tsukumi 2024-09-22 16:14:25 +09:00
• 9b27ba9777 Improve: Graphical display of ONNX conversion script logs tsukumi 2024-09-18 06:09:49 +09:00
• 449927616a Fix: Suppress many warning logs tsukumi 2024-09-18 05:48:06 +09:00
• fa7123c422 Fix: ONNX BERT model/tokenizer is not preloaded by default to avoid wasting VRAM tsukumi 2024-09-18 05:11:15 +09:00
• 327e038977 Improve: I/O binding during ONNX inference tsukumi 2024-09-18 04:49:56 +09:00
• 7d721fdf2b Fix: CUDAExecutionProvider was not being used to infer Style-Bert-VITS2 ONNX models even though CUDA was available tsukumi 2024-09-18 04:37:23 +09:00
• e5a05a30cb Improve: Outputs model load time and inference time tsukumi 2024-09-17 18:37:00 +09:00
• b95420325b Improve: Changed PyTorch to an optional dependency when used style_bert_vits2 as a library tsukumi 2024-09-17 17:56:59 +09:00
• 5e2c83c6c7 Improve: Support ONNX inference, add ONNX conversion script tsukumi 2024-09-17 16:55:03 +09:00
• f42b9f0f89 Fix: vocab.txt tsukumi 2024-09-17 13:44:21 +09:00
• 9de6e04ad7 Add: Preparation for ONNX inference support ④ tsukumi 2024-09-17 13:25:17 +09:00
• ebd4551d1e Add: Preparation for ONNX inference support ③ tsukumi 2024-09-17 09:16:16 +09:00
• b9fdeb4492 Add: Preparation for ONNX inference support ② tsukumi 2024-09-17 09:02:14 +09:00
• eede796c4e Add: Preparation for ONNX inference support ① tsukumi 2024-09-17 06:53:11 +09:00
• a037b955ac Refactor: run "hatch run style:fmt" tsukumi 2024-09-17 05:25:29 +09:00
• 7c5e839e82 Fix: Converts the given “N” phoneme to “n” when use_jp_extra=False tsukumi 2024-09-03 13:58:53 +09:00
• c140c0612a Fix: Old comment tsukumi 2024-09-02 01:49:49 +09:00