mrq/ai-voice-cloning
31
74
Fork 43
Code Issues 223 Pull Requests 5 Packages Projects Releases Wiki Activity
Collection of utilities aimed to voice clone through AI
356 Commits 1 Branch 0 Tags 12 MiB
Python 91.1%
Jupyter Notebook 4.9%
Shell 2.1%
Batchfile 1.1%
Dockerfile 0.8%
d16ba3a06d
Go to file
ben_mkiv d16ba3a06d Websocket server, override args parameters for model settings (squashed) 
favor existing arguments from parameters (kwargs) over global (args)

added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)
removed load_tts() call, the cli tool now relys on generate() to load the tts system, which is less fail prone imho

Revert "added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)"

This reverts commit d1dbe3e464.

Revert "added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)"

This reverts commit d1dbe3e464.

Revert "favor existing arguments from parameters (kwargs) over global (args)"

This reverts commit 89102347a9.

args are now updated in the websocket server

remove unused import
2023-08-23 19:27:21 +02:00
bin Initial refractor 2023-02-17 00:08:27 +00:00
models remove redundant phonemize for vall-e (oops), quantize all files and then phonemize all files for cope optimization, load alignment model once instead of for every transcription (speedup with whisperx) 2023-03-23 00:22:25 +00:00
modules forgot the important reason I even started working on AIVC again 2023-08-21 03:42:12 +00:00
results Initial refractor 2023-02-17 00:08:27 +00:00
src Websocket server, override args parameters for model settings (squashed) 2023-08-23 19:27:21 +02:00
training a bit of UI cleanup, import multiple audio files at once, actually shows progress when importing voices, hides audio metadata / latents if no generated settings are detected, preparing datasets shows its progress, saving a training YAML shows a message when done, training now works within the web UI, training output shows to web UI, provided notebook is cleaned up and uses a venv, etc. 2023-02-18 02:07:22 +00:00
voices Initial refractor 2023-02-17 00:08:27 +00:00
.dockerignore docker support 2023-04-07 21:52:18 +00:00
.gitignore experimental multi-gpu training (Linux only, because I can't into batch files) 2023-03-03 04:37:18 +00:00
.gitmodules while I'm breaking things, migrating dependencies to modules folder for tidiness 2023-03-09 04:03:57 +00:00
Dockerfile docker: add ffmpeg for whisper and general cleanup 2023-04-07 23:14:05 +00:00
LICENSE Initial refractor 2023-02-17 00:08:27 +00:00
notebook_colab.ipynb share if you 2023-03-08 15:59:16 +00:00
notebook_paperspace.ipynb fixed notebooks, provided paperspace notebook 2023-03-08 03:29:12 +00:00
README.md fixed issue with sliced audio being the wrong sample rate 2023-08-22 14:22:39 +00:00
requirements.txt made initialization faster if there's a lot of voice files (because glob fucking sucks), commiting changes buried on my training rig 2023-08-21 03:31:49 +00:00
setup-cuda-bnb.bat setup bnb on windows as needed 2023-03-16 20:48:48 +00:00
setup-cuda.bat DLAS is PIPified (but I'm still cloning it as a submodule to make updating it easier) 2023-03-21 15:46:53 +00:00
setup-cuda.sh DLAS is PIPified (but I'm still cloning it as a submodule to make updating it easier) 2023-03-21 15:46:53 +00:00
setup-directml.bat updated setup-directml.bat to not hard require torch version because it's updated to torch2 now 2023-04-29 00:50:16 +00:00
setup-docker.sh docker support 2023-04-07 21:52:18 +00:00
setup-rocm-bnb.sh while I'm breaking things, migrating dependencies to modules folder for tidiness 2023-03-09 04:03:57 +00:00
setup-rocm.sh DLAS is PIPified (but I'm still cloning it as a submodule to make updating it easier) 2023-03-21 15:46:53 +00:00
start-docker.sh docker support 2023-04-07 21:52:18 +00:00
start.bat added PYTHONUTF8 to start/train bats 2023-03-14 02:29:11 +00:00
start.sh :) 2023-03-14 05:02:14 +00:00
train-docker.sh docker: add training script 2023-04-07 23:15:13 +00:00
train.bat ;) 2023-03-14 15:48:09 +00:00
train.sh ;) 2023-03-14 15:48:09 +00:00
update-force.bat removed the hotfix pip installs that whisperx requires now that whisperx is gone 2023-03-11 16:55:19 +00:00
update-force.sh DLAS is PIPified (but I'm still cloning it as a submodule to make updating it easier) 2023-03-21 15:46:53 +00:00
update.bat added button to just load a training set's loss information, added installing broncotc/bitsandbytes-rocm when running setup-rocm.sh 2023-03-02 01:35:12 +00:00
update.sh added PYTHONUTF8 to start/train bats 2023-03-14 02:29:11 +00:00

README.md

AI Voice Cloning

Note This project has been in dire need of being rewritten from the ground up for some time. Apologies for any crust from my rather spaghetti code.

This repo/rentry aims to serve as both a foolproof guide for setting up AI voice cloning tools for legitimate, local use on Windows/Linux, as well as a stepping stone for anons that genuinely want to play around with TorToiSe.

>Ugh... why bother when I can just abuse 11.AI?

You're more than welcome to, but TorToiSe is shaping up to be a very promising tool, especially with finetuning now on the horizon.

This is not endorsed by neonbjb. I do not expect this to run into any ethical issues, as it seems (like me), this is mostly for making funny haha vidya characters say funny lines.

Documentation

Please consult the wiki for the documentation.

Bug Reporting

If you run into any problems, please refer to the issues you may encounter wiki page first.