~ft/rtmp

RTMP streaming for Plan 9
readme: update for mixfs
add -A option to specify audio offset (-200 for audio delayed by 200ms)

clone

read-only
https://git.sr.ht/~ft/rtmp
read/write
git@git.sr.ht:~ft/rtmp

You can also use your local clone with git send-email.

#rtmp

RTMP streaming for Plan 9. Tested with Twitch and PeerTube. Can stream to several RTMP urls at the same time.

#Usage (until I write a manpage)

video/hj264 -f 25 /dev/screen | video/rtmp -a ... rtmp://.... rtmp://...

For how to use rtmp with Twitch, refer to the documentation they provide. All you need is the correct RTMP URL. Preferably of a server that is close to you, for lower latency, see the list of ingest endpoints.

To stream with audio from the microphone, make sure it works, you can test it with pitch. Make sure audio/aacenc is installed. Then run this (change according to your preferences):

video/hj264 -f 25 /dev/screen | video/rtmp -a <{audio/aacenc -b </dev/audio} rtmp://....

Twitch does not require an audio stream to be present. PeerTube seems to require it, so in case you don't want to stream audio to PeerTube, it's possible to use silence by passing /dev/zero instead of /dev/audio.

For audio loopback you can do the following (make sure you have latest 9front):

audio/mixfs -s mixfs -m /n/mixfs
# this already binds on top of /dev/audio in that window

# in another window, provide silence so there are _always_ samples available for aacenc
mount /srv/mixfs /n/mixfs
dd -bs 8192 -if /dev/zero -of /n/mixfs/audio

# streaming, another window
mount /srv/mixfs /n/mixfs
video/hj264 -f 30 /dev/screen | video/rtmp -a <{audio/aacenc -b < /n/mixfs/audio} $url

You can also mix in audio from the microphone by mounting mixfs and writing PCM data from your phone to /n/mixfs/audio.