CouncilDataProject / speakerbox

Speakerbox: Fine-tune Audio Transformers for speaker identification.
https://councildataproject.org/speakerbox
MIT License
51 stars 6 forks source link

Adding new speakers #19

Open squeipom opened 1 year ago

squeipom commented 1 year ago

Hi!

First of all, thank you for developing SpeakerBox. It has proven to be an invaluable tool for my current project.

Allow me to provide you with some context regarding my situation. I am currently a third-year undergraduate computer science student, and I am attempting to transcribe all the meetings held among the individuals in my lab. To accomplish this task, I have been training the SpeakerBox model using a fixed set of approximately ten individuals from my lab.

However, I anticipate that new members may join the lab in the future, and it is crucial for me to be able to identify their voices as well. This is where I would greatly appreciate your guidance. Could you kindly offer any suggestions or recommendations on how I can tweak the SpeakerBox model to support the inclusion of new speakers?

I am particularly interested in understanding how I can adapt the model to incorporate the voices of new lab members seamlessly. Any advice or insights you can provide on this matter would be highly appreciated.

Thank you once again for creating SpeakerBox and for your time and assistance.