-
-
Notifications
You must be signed in to change notification settings - Fork 766
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Shared speaker embeddings across batch of audio streams. #1502
Comments
Thank you for your issue.You might want to check the FAQ if you haven't done so already. Feel free to close this issue if you found an answer in the FAQ. If your issue is a feature request, please read this first and update your request accordingly, if needed. If your issue is a bug report, please provide a minimum reproducible example as a link to a self-contained Google Colab notebook containing everthing needed to reproduce the bug:
Providing an MRE will increase your chance of getting an answer from the community (either maintainers or other power users). Companies relying on
|
Can you please clarify your use case? |
The microphones are attached to people who move around, so they have overlapped speech in periods, and separate in others. The microphones are synchronized. |
Sorry for the delay. I think we should sit together and talk for me to really understand and help you with your use case. |
Well, my hack works fine, so not much sense in paying for consulting there. This feature request was mostly about that the interface would be nicer if you could input a batch, and I believe it would be beneficial in a vast number of use cases, not just my own. However, I am interested in hearing about the premium models, so I'll send an email. |
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
Is your feature request related to a problem? Please describe.
I have a use case where i need to diarize a file containing multiple tracks, where each track corresponds to a separate microphone which all record concurrently. If I run each track individually through pyannote, the speaker IDs I cannot trust that speaker IDs correlates to the same person.
My current solution is to concatenate tracks, run them through pyannote as a single stream, and then fixing the speaker timings afterwards.
Describe the solution you'd like
The possibility to input a batch of streams. Each stream would be segmented individually, and then clustered using the same embedding space.
The text was updated successfully, but these errors were encountered: