Revisiting YourTTS - Details about Training, Datasets, and experiences Voice Cloning with Coqui TTS

Channel:
Canada NanoNomad
Subscribers:
2,550
Published on ● Video Link: https://www.youtube.com/watch?v=Yh7VGu5ZTXo


Duration: 22:47
1,418 views
28

Rambling about training YourTTS with the Coqui framework

This video is not about Coqui Studio, the paid web service. This is about Coqui TTS, the open source TTS framework.

I try the YourTTS model out of the box, then attempt fine tuning it. I gather some samples for datasets, try fine tuning the released model, it turns out like garbage, then I give up and retrain a better one from the beginning.

YourTTS training recipe from Coqui adapted for Colab:
https://colab.research.google.com/drive/1waQ1Y7odKCc75Y6Ujn3A5gRvndHlT_SJ?usp=sharing

Manual dataset processing pipeline:
https://github.com/visioninit/dataset-pipeline/tree/main

Download my multilingual, multispeaker YourTTS model on Huggingface: https://huggingface.co/AOLCDROM/YourTTS-Fr-En-De-Es
See allvoices.txt for information about each speaker:language training pair. Was trained on character sets, and uses 'artificial' language codes.

Semiautomated dataset tools:
https://github.com/rioharper/VocalForge

Other TTS videos:
https://youtu.be/8v18u8PQXgs
https://youtu.be/196h4JsqmZc
There's a whole playlist. Newest first for most recent scripts etc.

Re: Edresson's post referenced in the video - I can't find the link anymore. Google has like 4 results, whereas it had dozens of pages before.

https://tts.readthedocs.io/en/latest/tutorial_for_nervous_beginners.html

YourTTS release page with links to the paper and experiments:
https://github.com/Edresson/YourTTS



Other Videos By NanoNomad


2023-07-03.::Demo::. 4 Voice Multispeaker Tortoise TTS English Fine-Tuned Model Test :: Great Dictator Speech
2023-07-01Creepy Message about a 2003 Pandemic in China on found IBM PS/1 Pentium 66mhz PC
2023-06-28Now for Download: YourTTS (English, French, German, Spanish) Multilingual Model with 60+ Voices
2023-06-27Demo: YourTTS speaking in native French; A sampling of trained-in Voices
2023-06-27Demo: YourTTS speaking in native Spanish; A sampling of trained-in Voices
2023-06-27Demo: YourTTS speaking in native German; A sampling of trained-in Voices
2023-06-27Demo: YourTTS speaking Norman Arkawy's 1955 Sci-Fi Story 'Selling Point'. Info in description.
2023-06-14Running 13B and 30B LLMs at Home with KoboldCPP, AutoGPTQ, LLaMA.CPP/GGML
2023-06-08Demo and Download: YourTTS Multi-accent, English/Spanish Multi-Voice Model 600k Checkpoint
2023-06-05DEMO: YourTTS - One Voice, Many Accents. A single speaker can generate multiple accents.
2023-06-04Revisiting YourTTS - Details about Training, Datasets, and experiences Voice Cloning with Coqui TTS
2023-06-03DEMO: YourTTS Multi-speaker VCTK Irish-accented Dataset after 275k Steps trained using Coqui TTS
2023-05-22Tortoise TTS Fine Tuning Wrap-Up
2023-05-16Tortoise TTS DEMO: G-Man performs Gilbert and Sullivan's 'The Major-General's Song'
2023-05-15Train Tortoise TTS in English, Spanish, French, Italian, Portuguese, German, and more? Maybe?
2023-05-10DEMO: Testing French-Speaking Tortoise TTS
2023-05-10DEMO: Testing German-Speaking Tortoise TTS
2023-05-08DEMO: Testing Spanish Speaking Tortoise TTS
2023-05-07DEMO: Testing Tortoise TTS Speaking in Portuguese
2023-05-04Make Using Tortoise TTS Faster with Fine-Tuned Models
2023-05-01AI Voice Swap and Lip Sync using Wav2Lip-HQ-Updated


Tags:
YourTTS
TTS
Coqui
Text to Speech
AI Voice
AI Speech
voice cloning