Can speechBrain do batch inference for speaker embedding generation? #2886
Unanswered
sleepingcat4
asked this question in
Q&A
Replies: 1 comment 1 reply
-
|
Hey, not sure to exactly understand what you meant, but yeah, you should be able to provide in input to this function a sequence of shape |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
I actually have diagnosed the error and had reframed it as (BATCH, TIME) but then realised this wasn't extracting the speaker embeddings rather it was creating discrete embeddings for the entire audio. I have written this notebook, but I had another follow-up question:
|
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I was looking into this piece of code and curious if it could do batch inference for creating emebddings for 20-40 files at a time:
LINK TO GUIDE: https://speechbrain.readthedocs.io/en/latest/API/speechbrain.nnet.embedding.html
Beta Was this translation helpful? Give feedback.
All reactions