What is the maximum concurrency supported by mod_audio_fork?

[Chao Chen]

I would like to use FreeSWITCH's mod_audio_fork module in a BigBlueButton (BBB) conference to record each participant's audio stream into a separate file.

The goal is to extract each participant's speech and then generate a complete meeting summary based on all speakers' contributions.

When there are dozens of simultaneous participants or channels, I would like to know whether mod_audio_fork has any limitations on concurrent connections or audio streams.

[Marcel Hellkamp]

I'd wait for BBB 4.0 (or start testing against 4.0-beta releases) and tap into the LiveKIT streams. Developing something based on the soon-to-be-replaced freeswitch audio stack may be a waste of time.

Am 05.08.26 um 07:51 schrieb Chao Chen:

I would like to use FreeSWITCH's mod_audio_fork module in a BigBlueButton (BBB) conference to record each participant's audio stream into a separate file.

The goal is to extract each participant's speech and then generate a complete meeting summary based on all speakers' contributions.

When there are dozens of simultaneous participants or channels, I would like to know whether mod_audio_fork has any limitations on concurrent connections or audio streams.

--
You received this message because you are subscribed to the Google Groups "BigBlueButton-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bigbluebutton-...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/bigbluebutton-dev/921c545b-9ac4-4812-9dc6-75789cc539fan%40googlegroups.com.

[DistanceLearning.cloud]

thats a freeswitch issue on testing concurrency of module, i have in the past made and entry on dialplan to just write the raw audio of users wrie channel to a file, these files have the just the speaking audio of user before it is past into mod_conference and mixed.   you then map the freeswitch username to bbb username by mapping events.xml or listen real time ( in old versions redis, doent not apply)

Marcel is correct... unless you are pinned to freeswitch for your custom build, not the place to expend energy.  but gettins a prototype of per user speaking into a file, and then ran speech to text few lines of code.

There are some other opensource media servers that can listen to this streams, and ingest,  add a docker to server, and connect to whatever media server you happen to be running on.

Regards, 

Stephen