is it possible to mark vocals so ik could easy remove them?
I would like to remove all vocals in a file.
I know it is possible to mute the vocals but i want to complete remove them.
I know how to select and delete ⦠. Did that with a large file (2h) and thatās very time consuming (if you want to be precise, and that is what i want ) but i want to automate the process some more.
Let me be a little more precise.
I have big music files from 2h and larger. So not only one but multiple. And there are, spread over the file multiple vocal parts in it.
From those files i want to delete the vocal parts. Not by selecting it part for part but just like audacity is working with āVocal Reduction and Isolationā.
So it would be great if i could select a file and then activate a plugin like āeffect > Vocal Reduction and Isolationā which remove the parts with vocals in it.
That feature would be extremely difficult to develop. Personally I wouldnāt know where to start. Even the āstate of the artā speech recognition systems require voices to be reasonably clear with little other sound, so detecting vocals in music is likely to be a multi-million dollar project for one of the tech giants rather than for a tiny development team like Audacity.
That feature would be extremely difficult to develop.
Iām the first to agree with that !
But ā¦
Is the way āVocal Reduction and Isolationā is working not half of what Iām asking?
I mean ⦠if audacity is capable to locate the vocals and than āmuteā them, the positions (begin and end) are clear to audacity and that is actually just what i need.
From that point on itās easy for me to exactly isolate them and manually delete them.
That would be a perfect workaround for me.
But Iām not a developer and donāt know the amount of work that requires.
This is indeed a perfect workaround for me !
Locate the vocals and create a second track with only the vocals. In that way it is easy to locate them in the original track (above)
Or ⦠is āVocal Reduction and Isolationā (or other plugin) for me in some way workable ?
No. The way that vocal reduction works is to look for sound that is common in both left and right channels. Audacity has no idea about what that sound is. In a typical song recording there will be some degree of correlation between left and right channels throughout the song, whether there is a vocal present or not.
Yes, āVocal Reduction and Isolationā may be used to produce a āvocalā track as a guide to manually deleting vocals, but you will need to make the judgement calls about whether sound in the āvocalā track really is a voice or some other sound. How well āVocal Reduction and Isolationā (or āUnmixItā) works depends largely on the audio that you are working with. In some cases the separation can be fairly good, or in other cases it can be totally ineffective.
So ⦠if I will create an extra track with VRI.
Is it possible in one way or another to extract or compare this track with the original so there are (roughly) positions where to start manually ?
Apply Vocal Isolation / Reduction set to āIsolate Vocalsā to the duplicate tracks.
You āmayā be able to recognize some vocal regions just by looking at the processed track. You can also use the Solo / Mute buttons to listen to one track at a time (see: https://manual.audacityteam.org/man/audio_tracks.html#panel)
The duplicate track may be useful as a guide for editing the original track.