Spleeter: [Discussion] Separation of live audio from microphone
Hi,
I'd like to separate audio coming from the microphone, how do you suggest I should proceed. Is it possible? I'm currently creating small duration files and using them, but it doesn't seem like a clean approach.
Rahul-Sindhu
All 4 comments
Have you found a 'clean' solution for this?
markus-wa
on 5 Jan 2020
I do not think this is possible at the moment. The extended-abstract paper they published on ISMIR 2019 has a footnote at the end of the very first page that says:
[...] the model cannot be applied in real-time as it needs buffering
jsacrist
on 6 Jan 2020
Thanks for the clarification @jsacrist 馃檪
markus-wa
on 6 Jan 2020
this issue appears prominently in Google searches, so if you found this first like I did...
Check out vstSpleeter where despite the challenges @gvne implemented a real time VST interface to Spleeter! Also I've found @diracdeltas's popular fork of it helpful, particularly for the macOS build instructions.
more discussion and details on gvne's implementation in #276
varenc
on 10 Jun 2020
Related issues
rcgale
路
4Comments
JohnSmith2007
路
4Comments
NXG2005
路
4Comments
brunobulgaron
路
5Comments
geraldoramos
路
4Comments
Most helpful comment
this issue appears prominently in Google searches, so if you found this first like I did...
Check out vstSpleeter where despite the challenges @gvne implemented a real time VST interface to Spleeter! Also I've found @diracdeltas's popular fork of it helpful, particularly for the macOS build instructions.
more discussion and details on gvne's implementation in #276