noise-suppression-for-voice
noise-suppression-for-voice copied to clipboard
Question: Use as a filter for files
Hi,
it would be very nice to have the possibility to have a binary that can be used to process noise audio files and produce clean output files.
Personally I'd find it very useful to cleanup audio from lectures on Youtube.
Do you think is doable? Do you have any plans of implementing such a feature? If not, could you point me on how to do it?
Thanks
You can use the plugins in many audio/video editing programs.
For example, I use the VST plugin with Davinci Resolve.
Unfortunately I can't find a way to use this from the command line. Do you know of any software to do that?
I only use GUI based software, but maybe ffmpeg can transcode with the ladspa plugin.
FFmpeg also have native filter, arnndn, you need to get rnnoise model files though.
For the command-line tool. The original rnnoise repo may help you.
maybe ffmpeg can transcode with the ladspa plugin.
Yes, I think ffmpeg should be able to do this, see https://ffmpeg.org/ffmpeg-filters.html#ladspa
You might have to compile FFmpeg with LADSPA support.
You can use the plugins in many audio/video editing programs.
For example, I use the VST plugin with Davinci Resolve.
How? Most times it shows it loaded fine but I can't find it and resolve (Fairlight) is lame in it has no way to rescan your VSTs. When Resolve Studio 16 finally did show it for me it created a wide stereo and no reduction.
You can use the ladspa plugin directly with mpv:
mpv --af=lavfi='[ladspa=librnnoise_ladspa:p=noise_suppressor_mono]' file.mp4
This then should work with the youtube-dl integration of mpv.