Vocal Isolation

I want to keep the voice of the attached file and I fail at trying to do so.

I tried everything with hundreds of settings, but I just get either everything without the voice or just crap.

Any Ideas?

If you want to try, there’s the file. I am beyond desperation after 3 or 4 hours.

I can’t download the file.

Are you using Audacity 2.1.2? Are you using Effect > Vocal Reduction and Isolation?

Is it a mono file? It has to have two blue waves and they have to be different. Watch the bouncing sound meter when you play the work. Are the two meters exactly the same or bounce at exactly the same time? Then you have a two-track mono recording. The vocal tools don’t work on mono or any variation on mono.

Remove the sound clip and post it again.


That’s interesting. The forum claims it downloaded the file twice, but I didn’t get the file management tools or the actual file.


The file is mono (both channels are identical).
As the “analyze” option in “Vocal Reduction and Isolation” will tell you:

Pan position: 0
The left and right channels are correlated by about 100 %. This means:

  • The two channels are identical, i.e. dual mono.
    The center can’t be removed.
    Any remaining difference may be caused by lossy encoding.

I other words, it’s not going to work.
Separating vocals from that track is likely to require long and detailed editing with a program such as “Melodyne” (from $399 at time of writing)

So, if I can’t isolate the Voice, is there any kind of workaround?

@Kozikovski: DL worked for me. Still no file? :confused:

I have it now. It’s still mono. Audacity voice tools don’t work in mono. As above, there are special purpose programs for jobs like this, but they’re not easy to use or cheap.

It might be cheaper to get Will Arnett to read it for you live. Or write Warner Brothers. They may have the vocals in an isolated track. Lin Pictures or Animal Logic did the animation, they would certainly have the tracks for animation action and lip sync.


Warner Brothers? What are thex doing in the mix?

No offense please, but… it’s Max Payne 2 (which I assume you alredy figured out), so… no lip sync. Am I missing something here? :confused:

here is a hack of it: __

Apart from what has been already said, I note from your supplied sample, there are three major problems to get around here –

  1. The voice is very heavily processed, he’s been re-pitched with added sub-harmarnic overtones.
    Listen carefully to his voice, most phases down pitch to the end of the phase, and that is almost completely evenly done, not very human.

  2. The string instrument – cello? has a tonal range that overlaps the voice in many places – your extraction is good but in struggling to remove the cello you’ve altered the voice.

  3. Removing the siren is a tough job, it is not at one pitch but a number of glissando up and down across the voice. In removing this inharmonic noise you will inevitably knock out some hard consonants from the voice – like t’s, d’s, ess, etc.

Not a job for the fainthearted … so good luck!

Dave666 (the original poster) has not visited this forum since February, so I doubt that he will see your well wishes.
The post date is shown near the top left of each post.

Dave666 (the original poster) has not visited this forum since February, so I doubt that he will see your well wishes.
The post date is shown near the top left of each post.

I wrote that knowing that there would be quite a few others struggling with a similar problem and reading here.
To be honest my reply was for ANYONE and not just Dave666 attempting to remove the vocals from musical tracks.

Hai I want to create a karoke of the particular song which link is given below.There are two voices in the song.one is a women voice and other is a child voice.i am able to remove the women voice.but not able to remove the kids voice.please help me in this.the link which is given below is the song where women voice is removed.please help me in removing kids voice also .