Hey, so I have this idea of recording myself singing a song that I want lyrics for, having the AI transcribe it, and then pasting the lyrics into the original song.
When I try to do this through the lyrics editor, it just pastes the words, not the length of each one.
Is there a way I can access the data of a songs lyrics? is it like in vdjstems files?
When I try to do this through the lyrics editor, it just pastes the words, not the length of each one.
Is there a way I can access the data of a songs lyrics? is it like in vdjstems files?
geposted 6 hours ago
Unfortunately that info (lyrics, timings of when words should occur and duration/syllable breakdown if words) gets saved in a separate DB file and it's not readily editable, and the analysis needs some guiding vocals to know where to place words, etc (the editor doesn't allow word placement in terms of timings AFAIK).
It might be easier to sing a reference track on an instrumental, import it into VirtualDJ (to get that lyrics analysis), externally use AI to replace your vocals with another voice if necesary (and replace reference with the updated song - you might even be able to do the voice swap within VirtualDJ using the StemSwap sampler (swapping your vocals with the AI on)), then just stem out the vocals.
@devs AI voice replacement within VirtualDJ might actually be a nice follow-on new feature request - half the job is already done (lyrics + word positioning + reference voice in vocal stem for intonation).
It might be easier to sing a reference track on an instrumental, import it into VirtualDJ (to get that lyrics analysis), externally use AI to replace your vocals with another voice if necesary (and replace reference with the updated song - you might even be able to do the voice swap within VirtualDJ using the StemSwap sampler (swapping your vocals with the AI on)), then just stem out the vocals.
@devs AI voice replacement within VirtualDJ might actually be a nice follow-on new feature request - half the job is already done (lyrics + word positioning + reference voice in vocal stem for intonation).
geposted 5 hours ago
"Unfortunately that info (lyrics, timings of when words should occur and duration/syllable breakdown if words) gets saved in a separate DB file and it's not readily editable, and the analysis needs some guiding vocals to know where to place words, etc (the editor doesn't allow word placement in terms of timings AFAIK)."
I don't need to edit that DB file i could just swap the files. Where can I access those db files?
I don't need to edit that DB file i could just swap the files. Where can I access those db files?
geposted 44 minutes ago





