Multi-speaker TTS training #3448
-
|
Is there a Nemo recipe that supports multi-speaker TTS training using speaker embeddings? |
Beta Was this translation helpful? Give feedback.
Answered by
subhankar-ghosh
Aug 17, 2022
Replies: 2 comments
-
|
We are working on the feature. @subhankar-ghosh, could you please share guidance here? |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
Yes, we are working on this feature where we want to generate speaker embeddings using Speaker Verification model and use them to condition FastPitch. It is still Work in Progress. Thanks for the question. |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
subhankar-ghosh
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Yes, we are working on this feature where we want to generate speaker embeddings using Speaker Verification model and use them to condition FastPitch. It is still Work in Progress. Thanks for the question.