feat: add audio-only recording mode to reduce CPU usage

[agrille]

Summary

Adds a configurable audio-only recording mode that skips video capture and encoding entirely. This reduces CPU and storage requirements, making Jibri viable on lower-spec machines and ideal for use cases like meeting transcription.

Changes

• reference.conf: adds enable-video flag (default true) and audio-only FFmpeg input/output commands for Linux and macOS
• FfmpegCapturer: selects audio-only command when enable-video = false
• FileSink: uses .m4a extension for audio-only recordings
• Tests for both new code paths

Usage

Set in jibri.conf:
jibri.ffmpeg.enable-video = false

Output files use .m4a extension (AAC audio in MPEG-4 container).

Motivation

Recording a Jitsi meeting for transcription purposes does not require video.

[jitsi-jenkins]

Hi, thanks for your contribution!
If you haven't already done so, could you please make sure you sign our CLA (https://jitsi.org/icla for individuals and https://jitsi.org/ccla for corporations)? We would unfortunately be unable to merge your patch unless we have that piece :(.

[jjfreire]

+1

[mfmendez]

+1

[rodrigoGrZa]

+1

[Manuframil]

+1

[saghul]

Folks, just to adjust expectations: posting +1 won't help get this feature landed.

Specifically, the UI has no control over this, which means there is no way to decide whether one wants an audio-only or video recording.

For that reason alone, this cannot land.

If what you are interested in is transcriptions, the righ approach is to use JMR: https://github.com/jitsi/jitsi-multitrack-recorder which gets audio directly from the JVB and is much more lightweight.