Training Code

[fakerybakery]

Hi,
Thanks for releasing GLM-4-Voice! Are there any plans to release the training code?
Thanks!

[Opdoop]

If pre-train code can not be released, even fine-tuning code is helpful!
Please consider releasing the fine-tuning code!

[wtwong316]

I am working on an academic funded project related to the conversation using Cantonese dialect between 2 persons by finetuning ChatGLM 4. It is wonderful if the fine-tuning code of CLM-4-Voice model is released. Then I can extend my project not only text by also audio. It will be wonder if your team consider supporting academic research or even co-operation.

[darkacorn]

pong - now that the technical report is out - can we get some training examples ? even finetuning would do wonders

[AayushSameerShah]

I would want it so badly to train on a person's voice. Then the model can use the voice!!!

[darkacorn]

ya i given up on glm4voice . the model is weird - cant even tensor edit it to be abliterated .. the best bet is to wait on glm5voice and hope its open and actually somewhat better

[AayushSameerShah]

Yo @darkacorn ... is GLM4Voice trainable? I mean, you know...