StyleTTS2 - New Free Voice Cloning TTS #454

New Issue

FortermalGreek · 2023-11-26T16:54:48Z

FortermalGreek commented

2023-11-26 16:54:48 +00:00

This project looks promising

I wish there was a GUI or a tutorial for how to use it though.
(It was mainly built to read audio-books)

This project looks promising https://github.com/yl4579/StyleTTS2 I wish there was a GUI or a tutorial for how to use it though. (It was mainly built to read audio-books)

👍 1

DoctorPopi commented

2023-11-27 09:50:34 +00:00

To be followed indeed...!

👍 1

Bluebomber182 commented

2023-11-29 18:33:44 +00:00

You can use the Inference_LibriTTS.ipynb file locally
Just install these

pip install notebook jupyter

Then open the Inference_LibriTTS.ipynb file

jupyter notebook Inference_LibriTTS.ipynb

You can use the Inference_LibriTTS.ipynb file locally Just install these >pip install notebook jupyter Then open the Inference_LibriTTS.ipynb file >jupyter notebook Inference_LibriTTS.ipynb

👍 1

FortermalGreek commented

2023-11-30 05:04:21 +00:00

@Bluebomber182 Hey, I've actually seen the videos you've made on YouTube,!

But, I don't think you can fine tune your own model using the Inference files, thankfully someone made a guide on how to finetune our models https://github.com/IIEleven11/StyleTTS2FineTune

So I'm kinda confused, how did you clone these voices on YouTube? did you fine tune your own model? did you just do the One-Shot voice clone? or did you just use the included voice in StyleTTS2 and used your model of RVC with it?

@Bluebomber182 Hey, I've actually seen the videos you've made on YouTube,! But, I don't think you can fine tune your own model using the Inference files, thankfully someone made a guide on how to finetune our models https://github.com/IIEleven11/StyleTTS2FineTune So I'm kinda confused, how did you clone these voices on YouTube? did you fine tune your own model? did you just do the One-Shot voice clone? or did you just use the included voice in StyleTTS2 and used your model of RVC with it?

Bluebomber182 commented

2024-02-01 10:47:48 +00:00

@FortermalGreek
I finetune the voices by following these guides
https://github.com/yl4579/StyleTTS2/discussions/65#discussioncomment-7737796
https://gist.github.com/Shiro836/1c1881435ee75dc81069c505a14a9423
https://github.com/yl4579/StyleTTS2/discussions/81
I then put the styletts2 output files through rvc.

I also made a styletts2 tutorial. This doesn't include finetuning.
https://www.youtube.com/watch?v=2dg_xRnMYT4

@FortermalGreek I finetune the voices by following these guides https://github.com/yl4579/StyleTTS2/discussions/65#discussioncomment-7737796 https://gist.github.com/Shiro836/1c1881435ee75dc81069c505a14a9423 https://github.com/yl4579/StyleTTS2/discussions/81 I then put the styletts2 output files through rvc. I also made a styletts2 tutorial. This doesn't include finetuning. https://www.youtube.com/watch?v=2dg_xRnMYT4

Bluebomber182 commented

2024-04-27 19:29:41 +00:00

I found an improved styletts2 base model for those interested in fine tuning.
https://huggingface.co/ShoukanLabs/Vokan

I found an improved styletts2 base model for those interested in fine tuning. https://huggingface.co/ShoukanLabs/Vokan

Sign in to join this conversation.