Question about the SRT-protocol and videos with multiple audio tracks

[Sanderluc5]

Hello,

I have a question about using FFMPEG with SRT and working with streams that have multiple audiotracks.

Can using FFMPEG with SRT provide advantages over RTMP for live streaming to clients? The documentation suggests SRT may perform better, but I'm uncertain if this benefit remains when SRS converts the stream to RTMP. I currently use HTTP-TS for live streaming to my customers and am seeking the highest performance possible. Would SRT contribute to this goal?

If SRT is utilized, and SRS subsequently converts the stream to RTMP before remuxing to HTTP-TS, could this process introduce extra latency? Directly using FFMPEG with RTMP might bypass this additional conversion step.

I have another question as well. I also work with videos that have multiple audio tracks. While MPEG2TS supports multiple audio tracks, RTMP does not. Are there any workarounds to deliver the SRT stream in an HTTP-TS format without the need of RTMP directly, possibly with transcoding?

TRANS_BY_GPT4