AAC | ADPCM | Convolution | DAFx | FFT | IIR | Mixer | MP3 | MPEG | MPEG-4

Anybody please tell me by what means, a voice can be removed from an audio
signal.

Is this just an application of band pass filter on the actual signal or
completely different approach?

______________________________
DSPRelated.com's 50,000th member announced! Details Here.



(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )

have a look at this
http://www.aes.org/e-lib/browse.cfm?elib=14527

On Wed, May 27, 2009 at 1:02 AM, jr_vignesh <r...@gmail.com> wrote:

> Anybody please tell me by what means, a voice can be removed from an audio
> signal.
>
> Is this just an application of band pass filter on the actual signal or
> completely different approach?
>
>  
>

-- 
Regards
Shamail

(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )

I would imagine you could remove some aspects of the voice signal
through band reject filters. By far the best way, if you have access to
the vocal track or acapella is to to phase invert the vocal, and then
mix with the original. This results in a completely clean and full range
instrumental track - but this method is only useful if you have the
vocal track.

Maybe some kind of formant reject? Sorry if that was useless!

Tom.

jr_vignesh wrote:
> Anybody please tell me by what means, a voice can be removed from an
> audio signal.
>
> Is this just an application of band pass filter on the actual signal
> or completely different approach?

(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )

JR-

> > Anybody please tell me by what means, a voice can be removed from an
> > audio signal.
> >
> > Is this just an application of band pass filter on the actual signal
> > or completely different approach?
>
> I would imagine you could remove some aspects of the voice signal
> through band reject filters. By far the best way, if you have access to
> the vocal track or acapella is to to phase invert the vocal, and then
> mix with the original. This results in a completely clean and full range
> instrumental track - but this method is only useful if you have the
> vocal track.
> 
> Maybe some kind of formant reject? Sorry if that was useless!

If you don't have the original vocal signal, then filtering is mostly useless
--
frequency overlap between lead vocal and music content will be in many cases
indistinguishable.  One common technique in the music / performing arts business
is
to work with stereo recordings -- the assumption is the lead vocal was recorded
in
the middle, most instrumentals and backup vocals are on one side or the other,
so L/R
channels can be subtracted to reduce the lead vocal.  You might try searching
Google
for 'lead vocal reduction'.

I think some of the popular music editing software, such as CoolEdit, GoldWave,
and
YoGen, include vocal reduction algorithms based on this method.  I've not seen
inexpensive off-the-shelf that purports to reduce lead vocal on a mono
recording. 
Such an algorithm would be adaptive and substantially more complex.  For that
you
might search professional products, such as Superscope Technologies. 

-Jeff

______________________________
New TMS320C5514/15 DSPs - Industry's lowest power 16-bit DSP platform extends with two new devices. Details Here!



(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )