[music-dsp] Extract voice from a stereo signal

[Rory Walsh]

You may want to take a look at the following paper from the DAFX 04 
conference proceedings.

http://dafx04.na.infn.it/WebProc/Proc/P_240.pdf

If you happen to use Csound at all I know that one of the Csound guys is 
after writing a Csound opcode that does the trick. I have only heard it 
work with a jazz quartet and not voice, but it seems to work well and 
isolating the double bass for example ain't too tricky. Saying that I'm 
no expert on these matters!



Thomas Rehaag wrote:
> Hi,
> 
> 
>>original signal - signal with voice removed = voice
> 
> 
> IMHO this doesn't work.
> 
> No voice mono = left - right
> left - (No voice mono) = left - (left - right) = right
> right - (No voice mono) = right - (left - right) = 2 * right - left
> 
> There is no mixture of these simple equations that will deliver
> 'middle'.
> 
> Cheers,
> 
> Thomas
> 
> Florian Schmidt schrieb:
> 
>>On Wed, 23 Nov 2005 16:28:50 +0000
>>Avelino Herrera <avelinoherrera at hotpop.com> wrote:
>>
>>
>>>Hi all.
>>>There is a lot of voice removal algorithms mainly based upon the
>>>substraction of the left and right channel signals. But, are there any
>>>voice extraction algorithm? I want to obtain only the voice of a stereo
>>>song assuming that the voice are located at center of the stereo space.
>>>Thanks in advance.
>>
>>original signal - signal with voice removed = voice
>>
>>same pitfalls as to voice removal apply.
>>
>>Flo
>>
>>--
>>Palimm Palimm!
>>http://tapas.affenbande.org
>>--
>>dupswapdrop -- the music-dsp mailing list and website:
>>subscription info, FAQ, source code archive, list archive, book reviews, dsp links
>>http://shoko.calarts.edu/musicdsp
>>http://ceait.calarts.edu/mailman/listinfo/music-dsp
> 
> --
> dupswapdrop -- the music-dsp mailing list and website: 
> subscription info, FAQ, source code archive, list archive, book reviews, dsp links 
> http://shoko.calarts.edu/musicdsp 
> http://ceait.calarts.edu/mailman/listinfo/music-dsp
>