gitea-mirror/F5-TTS
1
0
Fork 0
mirror of https://github.com/SWivid/F5-TTS.git synced 2025-12-25 12:24:54 -08:00
Code Issues Packages Projects Releases Wiki Activity
520 Commits 1 Branch 15 Tags
621559cbbea901b677d48c6bb8ac88366a74fd12
Commit Graph

42 Commits

Author SHA1 Message Date
SWivid
621559cbbe v1.0.7 2025-03-21 14:40:52 +08:00
SWivid
c6b3189bbd v1.0.6 improves docker usage 2025-03-20 22:48:36 +08:00
SWivid
a1e88c2a9e v1.0.5 update finetune_gradio.py for clearer guidance 2025-03-17 21:50:50 +08:00
SWivid
1ab90505a4 v1.0.4 fix finetune_gradio.py vocab extend with .safetensors ckpt 2025-03-17 16:22:26 +08:00
SWivid
7e4985ca56 v1.0.3 fix api.py 2025-03-17 02:39:20 +08:00
SWivid
f05ceda4cb v1.0.2 fix: torch.utils.checkpoint.checkpoint add use_reentrant=False 2025-03-15 16:34:32 +08:00
Yushen CHEN
297755fac3 v1.0.1 VRAM usage management #851 2025-03-14 17:31:44 +08:00
SWivid
ca6e49adaa 1.0.0 F5-TTS v1 base model with better training and inference performance 2025-03-12 17:23:10 +08:00
SWivid
09b478b7d7 0.6.2 support socket_server.py with general text chunk 2025-02-25 04:47:40 +08:00
SWivid
a72f2f8efb 0.6.1 fix tqdm func check with difference call behavior from gr.Progress() 2025-02-22 08:33:10 +08:00
SWivid
7ee55d773c formatting 2025-02-21 17:00:51 +08:00
rino
4ceba6dc24 This patch is to solve a problem where streaming will handle all of the client input 
[add] numpy tokenizer for stream chunk
[add] infer_batch_process_stream in utils_infer
[add] file writter after streaming

[edit] adjustment for streaming server
[edit] data handling processes and sends chunk by chunk
[delete] threading on processing the inference, just for file writting
 2025-02-21 14:35:01 +07:00
SWivid
5fa0479432 0.5.3 fix MPS device compatibility; update readme 2025-02-18 18:42:03 +08:00
SWivid
f062403353 0.5.2 Improve prepare_csv_wavs.py from @hcsolakoglu 2025-02-09 14:36:40 +08:00
SWivid
261b2774f2 0.5.1 Enhance DynamicBatchSampler to support epoch-based shuffling 2025-02-05 15:12:34 +08:00
unknown
f0996492a7 0.5.0 fix grad_accum bug from 0.4.0, #715 #728 2025-01-29 15:18:02 +08:00
unknown
0d95df4a4d 0.4.6 minor fixes for finetune-gradio -cli 2025-01-29 00:06:10 +08:00
unknown
607b92b391 0.4.5 fix extremely short case that lengths of text_seq > audio_seq, causing wrong cond_mask 2025-01-28 12:38:16 +08:00
unknown
ee2b77064e 0.4.4 fix hard coded stdout for finetune-gradio gui 2025-01-28 11:39:54 +08:00
unknown
d1f708d442 0.4.3 Bug-fixes for finetune-gradio component mismatch & checkpoint loading error loop 2025-01-27 21:22:29 +08:00
SWivid
9e51878d18 0.4.2 fix trainer with grad_accum 2025-01-15 18:28:41 +08:00
SWivid
12d6970271 0.4.1 #718 add keep_last_n_checkpoints option 2025-01-15 15:06:55 +08:00
unknown
0b11f7eae6 0.4.0 fix gradient accumulation; change checkpointing logic to per_updates 2025-01-12 21:26:57 +08:00
SWivid
3e73553bd9 v0.3.4 2024-12-22 11:09:29 +08:00
SWivid
6ab873fb19 Fixed #658 2024-12-21 23:19:11 +08:00
SWivid
deaca8d24c v0.3.2 add flags and default values to socket_server.py 2024-12-18 20:32:20 +08:00
SWivid
84978268f0 v0.3.1 2024-12-17 07:59:59 +08:00
SWivid
61f28ee8a5 v0.3.0 custom model cfg, checkpointing for training, minor fix, etc. 2024-12-16 16:39:56 +08:00
SWivid
299f0aa8bc v0.2.1 Fixed #545 2024-11-28 12:36:45 +08:00
SWivid
771007b462 v0.2.0. hydra config for training 2024-11-28 01:28:38 +08:00
ZhikangNiu
65a649c683 fix minor bug and rename config -> configs 2024-11-28 00:24:33 +08:00
SWivid
194bf1e853 formatting 2024-11-18 22:33:13 +08:00
SWivid
cb8ce3306d update. compatibility with mps device #477 thanks to @aboutmydreams 2024-11-17 18:57:28 +08:00
SWivid
a23ec25b39 v0.1.1 2024-11-11 11:19:04 +08:00
SWivid
dcd9a19889 v0.1.0. Add custom model support for local deploy; add share model cards, etc. 2024-11-09 04:35:44 +08:00
cocktailpeanut
6a104b4025 add train interface (#258) 
* add train interface
* Update README.md
* Update pyproject.toml
 2024-10-25 13:44:38 +08:00
Hailey Collet
86e1e1e9b8 Update pyproject.toml - Specify required Gradio version 
Specify required Gradio version to avoid errors with the Progress object. Earlier versions will give the error:
```
File ".../F5-TTS/model/utils_infer.py", line 268, in infer_batch_process
    text_list = [ref_text + gen_text]
TypeError: can only concatenate str (not "Progress") to str
```
I am pretty sure the PR which changed this behavior is https://github.com/gradio-app/gradio/pull/5693, which was merged for 3.45.2. Certainly, you get the above error with 3.36.1 and it works with 3.45.2.
 2024-10-24 12:20:38 -06:00
SWivid
8629c6f91f initial updates for infer stuffs 2024-10-24 23:51:20 +08:00
SWivid
ba4b04ba55 finish eval dependencies; update infer_gradio with chat feature 2024-10-24 18:39:02 +08:00
SWivid
254e5e6d30 update finetune-cli -gradio 2024-10-24 15:23:55 +08:00
SWivid
d8638a6c32 . 2024-10-23 23:05:25 +08:00
Yushen CHEN
c4eee0f96b convert to pkg, reorganize repo (#228) 
* group files in f5_tts directory

* add setup.py

* use global imports

* simplify demo

* add install directions for library mode

* fix old huggingface_hub version constraint

* move finetune to package

* change imports to f5_tts.model

* bump version

* fix bad merge

* Update inference-cli.py

* fix HF space

* reformat

* fix utils.py vocab.txt import

* fix format

* adapt README for f5_tts package structure

* simplify app.py

* add gradio.Dockerfile and workflow

* refactored for pyproject.toml

* refactored for pyproject.toml

* added in reference to packaged files

* use fork for testing docker image

* added in reference to packaged files

* minor tweaks

* fixed inference-cli.toml path

* fixed inference-cli.toml path

* fixed inference-cli.toml path

* fixed inference-cli.toml path

* refactor eval_infer_batch.py

* fix typo

* added eval_infer_batch to scripts

---------

Co-authored-by: Roberts Slisans <rsxdalv@gmail.com>
Co-authored-by: Adam Kessel <adam@rosi-kessel.org>
Co-authored-by: Roberts Slisans <roberts.slisans@gmail.com>
 2024-10-23 21:07:59 +08:00