How can I get audio streaming realtime?

Hello,

I’ve used jitsi-meet and jibri for video conference working well with a recording session. Now, I’ve another feature which required audio streaming for some AI operations I want to do.
How can I retrieve real-time audio streaming? I want to give an API where they can get a real-time stream of audio, and do the further operation. Please guide me how can we do it?

We had added transcribe feature to jigasi. Jigasi enters the conference and sends the stream to google transcriber service and returns back the result in form of subtitles to participants, sounds the same thing you want to achieve.
The transcribe service is also abstracted so you can easily plug your own implementation.


Thank you for the answer.

Correct me If I’m wrong,
If I want to fetch continues streaming and need to call another API then I need to be update following file. Right?

You need to implement TranscriptionService, put it in a jar with all dependencies, put it in jigasi lib folder, add in jigasi sip-communicxator.properties file org.jitsi.jigasi.transcription.customService= with the full class name of your implementation and that’s it. You will replace jigasi’s translation service with yours.

Thank you for the response.

So, there is no way to get streaming via APIs. everything we need to build in jigasi as custom TranscriptionService. Right?

Yes.