Hi all! New to this Caspar world and have a quick question as I’m pretty new to this very powerful tool. We’re actually looking at using this as a backend in some of our normal switching rigs (physical and cloud) but the last part I’m not sure if it can do, so any help is appreciated.
We’re running OBS Ninja to pull in ISO cams of speakers into our server and putting them our separate channels, then we’d like these separate channels to be output NDI into our rigs so our ops can switch. I’ve got it all working, pulling us OBS ninja as HTML (similar to playing a youtube video) but I’m not getting any sound.
I do get sound on other playback sources (videos etc) so the routing is good (I think)
Tried this on a few server versions, 2.0, 2.1, 2.3, NRK, just wondering if anyone has experience pulling audio through HTML and had NDI on the same server.
Thanks in advance!
Playing audio that way does not work, it goes to the PC speakers. I would try VLC player to pull the streams and let it output to NDI. There is a plugin from the Newtek website to enable NDI in VLC.
Thank you for the advice! Unfortunately VLC can’t play back that particular website (Basically a google meet, but we’re able to pull in individual streams for each participant in the call). Is there any way to route PC speaker audio to go to one of the Consumers? That would work in our situation (I Consumer with the entire audio mix and separate Consumers for each participant channel
In my opinion you are using wrong tool.
OBS can input the HLS-stream from OBS.ninja and output it as SDI or NDI, and you can start multiple instaces of OBS on one machine.
@markunsygar yes I’ve been pulling these in with obs (actually using the NDI plugin) and enabling NDI output on each browser source (1 per participant). It works, and it works fairly well I’m just finding the CPU load and GPU load of the machine is higher than when I was using Caspar (without audio)
Not a real “official” way. Probably you can try to use NDI screen capture tool to send the desktop with audio to Caspar or directly to your NDI consuming system. I don’t know what that does to the lip sync, but it’s probably worth a try.
Routing audio from the HTML producer as not been implemented yet as it isnt possible in the version of CEF(chromium) we are using.
It is on the todo list for once that is updated, but that is currently blocked on a different issue of newer versions not supporting gpu acceleration very well
1 Like