How to add a custom AudioSource?

[haydenkaizeta]

Hello all,

I am looking to add a custom audio source that will provide audio data from somewhere else . I have already forked the (this) KMP library but it seems like I would need to go into the native SDK in order to do this. Does anyone have an idea on how to do this?

The ideal implementation would allow me to take a Flow of ByteArray (of PCM or similar) data and send it over the webrtc audio channel

Thanks

[shepeliev]

Hello @haydenkaizeta,

Unfortunately, WebRTC KMP doesn't have any API for injecting a custom AudioSource. So you're right, you have to use the native SDK to do this.

[haydenkaizeta]

@shepeliev yeah that makes sense but do you have any thoughts on the underlying API being able to do this in Java/Kotlin? Let's constrain it to Android for now. If I could get that working in a fork I'd be happy to push it back into the repo

[shepeliev]

I'd start by researching WebRTC sources to understand how they create AudioSource, then attempt to implement a custom one. However, I assume this task would require a custom WebRTC SDK build that exposes an API for audio source customization.

[haydenkaizeta]

@shepeliev many thanks! I'll update here if we make progress