How to FineTune the model on New Language?
#28
by
alvynabranches
- opened
I have around 75k sentence pairs of English and Konkani. I want to train this model on those text sentence pairs.
Also can we do zero shot for text to speech or speech to text (English input speech and Konkani output text) by just using text to text fine-tuning?
alvynabranches
changed discussion title from
How to FineTune the model on New Language.
to How to FineTune the model on New Language?
Text to Text finetuning is not available for Seamless m4T. You can check out ASR, S2ST, and S2TT here in their Github Repository: https://github.com/facebookresearch/seamless_communication/tree/main/src/seamless_communication/cli/m4t/finetune
Finetuning might not allow Zero-Shotting other data combinations.