obs-web VS rnnoise

OBS-Web remote control - Control OBS remotely from a web browser (on the same LAN)

Niek obs-web http://obs-web.niek.tv/ works great combined with noalbs or bot commands for manual scene controls

I use http://obs-web.niek.tv/ connected to OBS Websockets to control scene switching from my Galaxy Watch 5 Pro. If the Apple Watch has a reasonable web browser, you may be able to do the same. This really is designed for a mobile phone or tablet, but works fine on my watch.

OBS v28 is amazing. It's bundled with the obs-websocket v5 plugin now. I have updated OBS-web [1] to be compatible with this, so you can now control OBS remotely without any additional setup.

1. https://github.com/Niek/obs-web

Another option is minimizing OBS and using https://github.com/Niek/obs-web to control OBS from a tablet or phone.

Self-promo:

In case anyone is interested in controlling OBS remotely, you can use OBS-web: https://github.com/Niek/obs-web

It's not a full reimplementation of the UI, but offers the most often used controls so you can manage a livestream from e.g. a tablet or phone.

About how the director is controlling the scenes. I've one assistant connected on the GC VM via remote desktop to monitor the transmission and do the audio cues. For the director I've obs-websocket set up and the director is connecting through [obs-web](https://github.com/Niek/obs-web). I am also an actor in this play. Not 100% happy with this setup though. Perhaps the best would be the assistant in a zoom call with with the director and changing the scenes for her as commanded.

It has been done before, for eg the original RNNoise was trained on proprietary data, later there was crowd-sourced effort to record new data and have it under libre licenses.

https://github.com/xiph/rnnoise/

This reminds me a lot of https://github.com/xiph/rnnoise and my use of it locally. It zeroes in on voice via RNN which seems to beat most other noise detection filters I've tried. Unfortunately, I mostly disable it these days since it's a bit harder to tune than I'm up for, but it's by far the most promising local noise reduction I've used.

Sure.

Carla: https://github.com/falkTX/Carla

It lets me install any normal audio pro audio plugins, for example https://github.com/xiph/rnnoise

It also does some cable management, but qpwgraph is maybe better for that.

I looked at your code and the approach (IMO) is kind of bad.

If you want to solve the problem of "voice changer", you can skip the UI entirely and just use plugin parameters. You can also skip the problem of managing the connections. And when you publish your work, every pro audio software (Ableton, Reaper, whatever) can use your audio processing.

Hope that helps.

Interesting! How does it compare with NoiseTorch/RNNoise?

Das ist eine open source lib. Damit sollte das klappen. https://github.com/xiph/rnnoise/blob/master/examples/rnnoise_demo.c

RNNoise (behaves similarly to RTX broadcast/voice/whatever the fuck they're calling it now, but with significantly better performance) - plugs into OBS or other programs flawlessly

OBS ships with rnnoise noise reduction, which is like NVIDIA Broadcast, but works on any CPU. See also NoiseTorch and EasyEffects if you're on Linux.

It's pretty great, works decently, but the sad thing is the author put it out a few years ago, wrote a paper and then moved onto something else and it's pretty much unmaintained and requires some very specific ML knowledge.

https://github.com/xiph/rnnoise