keep voice remove music..reverse karaoke?

[Adam Rose esq.]

good luck! people have enough difficulty removing noise or glitches – removing entire backing sound ain’t gonna be easy

😉

[Mike Kujbida]

[Fuddam] “removing entire backing sound ain’t gonna be easy

;)”

The word “impossible” comes to mind whenever I hear this request 🙁
Find a decent singer,record him/her singing the song and place that on a separate track to use as the reference for your lip sync software.
Once everything is lined up to your satisfaction, remove your new vocal track so that all you have left is the original song.

[Terry Nahirny]

Mike has good advice with recording a new vocal track, however, if that’s not possible….

you can run the song through an equalizer and remove the low and high frequencies and boost the mid frequency of the voice. Depending on what is in the music you should be able boost the vocal frequencies and reduce the unwanted music frequencies. This may or may not work well enough and depends again on the music in the song.

it’s worth a try for sure -works for me in the studio here to a point. whether it’ll work well enough for your lip sinc software….??

t.

[Bodybuildingbs]

Thank you all for your advice..the removal does not have to be perfect as I am not going to use it for the final, I will use the original song..just enough so that the software doesnt convert the drum beat to speech etc..actually the removed vocal can sound horrible…

[Terje A. bergesen]

You don’t need to find a “decent” singer, if you can manage it, download the lyrics etc, and record the lyrics your self on a new track. Do the lip synch to your voice. You can remove your voice from the final production if you are able to lip synch close enough to the original song.

Now the lip synch software is synched to your voice, but is playing to the original song. Should work fine since the quality of your voice won’t matter all that much when it comes to the technical part of doing the actual lip synch.

[Terry Esslinger]

Actually I think that you could put on a set of headphones and play the song. Then sing it yourself and record your voice. You have the headphones so that you can sing the lyrics at the right time (as per the original sound). Then lipsynch your project to your voice recording. Then totally replace your voice recording with the original sound. Should work.

[Mike Kujbida]

To Terje & Terry, excellent suggestion about the OP using his own voice since all that’s needed is a voice and not a “decent” one.
Not to say that Jack doesn’t have a good voice as I have no idea 🙂

[Arthur Bueno]

Actually I saw a plugin once that can either isolate or remove the voice from a song (sort of). It’s called ‘voice trap’, written by Trevor Magnusson, and you can download it at https://www.cloneensemble.com/. In a perfect world it would work seamlessly together with your lipsync program (but I also think singing along is the smart down to earth solution here).

[Trevor Asquerthian]

I have had some success in the past, when dealing with a mix (inc voice) and m&e track (i.e. mix without voice) in isolating the voice only.

Mix the full mix together with the phase inverted m&e – thus where the content of the two sources is the same they cancel each other (because of the phase inversion) and where they are different there is no effect (i.e. on the voice).

This does require that:
a. the mix=m&e (no extra compression, eq, delay etc)
b. you actually have the m&e (which may be more of a problem).