How to isolate user's voice for transcription with background speakers

Agents
1

This question originally came up in our Slack community and the thread has been consolidated here for long-term reference.

What’s the most effective way for user voice isolation, capturing only the intended user’s voice as part of the transcription?

I’m trying to solve for cases where there are multiple people talking in the background.

2

You could try using Background Voice Cancellation (BVC):

BVC is designed to isolate the primary speaker’s voice and reduce background voices.