Full-duplex voice call with Cloudflare SFU?
Hi Guys,
I’m working on building a voice call agent using Cloudflare SFU Realtime. My setup is:
The client runs in the browser and connects over WebRTC.
On the server side, I’m using Cloudflare to establish a headless WebRTC connection.
I’m using stream adapters to capture the client’s audio → run it through AI processing → then send the audio back to the client using an ingest adapter.
Essentially, I’m trying to establish a full-duplex bidirectional voice call where the client and AI can talk to each other in real time.
I wanted to check if there’s a proper solution or best practice for this workflow. Is there any example repository, reference implementation, or documentation that demonstrates this kind of setup
Thanks
0 Replies