resemble-ai
Training Code?
First off, just want to say a heartfelt thanks for all the love and excitement around this model. We’re genuinely grateful for the open source community, and honestly, a lot of what we do is inspired by your projects, questions, and experiments.
We’ve had quite a few people asking for the training code. I get it, openness is a big part of what makes this space awesome. I want to be real with you: building these models is not cheap. Training, testing, and keeping everything running takes a lot of resources, both time and money. In order to keep this going and actually be able to release more cool stuff in the future, we need a way to keep the project sustainable.
For now, that means we’re not releasing the training code, and fine-tuning will be something we support through our paid API (https://app.resemble.ai). This helps us pay the bills and keep pushing out models that (hopefully) benefit everyone.
We love being part of this community, and we’re doing our best to strike a balance between openness and being able to keep doing the work. We hope you understand. If you have thoughts, feedback, or just want to chat, our door is always open.
https://github.com/stlohrey/chatterbox-finetuning code works fine, I am using it on some of mine private datasets - not sure about the default hyperparameters and the dataset size needed to train the new language...
There are nice reports about training on dataset in German language - https://wandb.ai/sebastian-bodza/t3-training/reports/Chatterbox-Finetune-600k-data--VmlldzoxMzE1Mzg5MA and https://x.com/SebastianB929/status/1932333058616631522 - however for my dataset i needed to use lower lr rates because even with long warmup phase the training was still unstable with such a high lr rate...
I speak German so I can tell that the quality of the model is good, but I haven't done any massive testing so I am not sure about the stability of the output like word error rates, etc.
@C00reNUT ,
I am working with Indian languages. The fine-tuned model is generating audio, but the speech does not correspond to the input text. How did you handle tokenization for German?
