Training data information

#3
by jpgallegoar - opened

Thank you for releasing this model, could you give us any insights into how many hours of audio it took to train?

HKUST Audio org

250,000 hours of mixed Chinese and English speech data

Thank you, is it possible to finetune this model?

Thank you, is it possible to finetune this model?

There is this - not sure of the relationship to the og authors

https://github.com/zhenye234/LLaSA_training

Sign up or log in to comment