publish to WebRTC and read from LL-HLS still has mono opus audio

[null4n]

• [x] Linux amd64 standard
• [ ] Linux amd64 Docker
• [ ] Linux arm64 standard
• [ ] Linux arm64 Docker
• [ ] Linux arm7 standard
• [ ] Linux arm7 Docker
• [ ] Linux arm6 standard
• [ ] Linux arm6 Docker
• [ ] Windows amd64 standard
• [ ] Windows amd64 Docker (WSL backend)
• [ ] macOS amd64 standard
• [ ] macOS amd64 Docker
• [ ] Other (please describe)

Describe the issue

Using lasted chrome to publish webcam and mic on http://localhost:8889/stream/publish Video codec choose AV1 and audio is Opus, uncheck "optimize for voice" When I read read from ffmpeg using HLS, it shows audio is mono. Using default .yml config file in ffprobe Stream #0:1: Audio: opus (Opus / 0x7375704F), 48000 Hz, mono, fltp, 0 kb/s (default)

Describe how to replicate the issue

1. start the server
2. publish with chrome
3. read with ffprobe http://localhost:8888/stream/index.m3u8

Did you attach the server logs?

no

Did you attach a network dump?

no

[zapf2000]

It is also possible to stream to the server with vdo.ninja. There you have a lot more parameters you can choose from. https://vdo.ninja/whip

[aler9]

Thanks for reporting the issue, this is fixed with #3263

[github-actions[bot]]

This issue is mentioned in release v1.8.0 🚀 Check out the entire changelog by clicking here