Is there a way to finetune the Seamless-m4t-large model for Text-to-Text task, as the official git repo "facebookresearch/seamless_communication" doesn't support the same currently? Worked on modifying the repo code for the same but did not find any leads. Continue reading...
