Skip to content
Discussion options

You must be logged in to vote

Not out-of-the-box, because the probability distribution of sampling timestamps will be different between the translation and transcription modes. It might be possible to hack a pipeline of transcribing with word timestamps and enforce the positions of the timestamp tokens in the translation mode (by editing decoding.py), but it may be easier to just use a separate translation API at that point..

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by jongwook
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants