Speex dev - Aug 2004 - reduction of noise due to high microphone gain

This works really well for white noise reduction.  However what I've noticed
was the amplitudes of normal speech samples also get reduced.

Is this something by design, or is there a way to automatically recover the
original speech sample volumes ?

<p>Thanks.

<p>Tongbiao

<p>-----Original Message-----
From: Jean-Marc Valin [mailto:jean-marc.valin@hermes.usherb.ca] 
Sent: Tuesday, August 12, 2003 12:44 PM
To: speex
Cc: Tongbiao Li
Subject: Re: [speex-dev] reduction of noise due to high microphone gain

Hi,

You might want to try the denoiser I've got in CVS (that will go into
Speex 1.1 eventually). It will remove stationary noise from the signal
and probably help the silence detection. See testdenoise.c for how to
use it. There's also a better VAD in the denoiser but speex_encode
cannot make use of it yet.

        Jean-Marc

Le mar 12/08/2003 à 15:03, Tongbiao Li a écrit :
> Hello,
> 
>  
> 
> With high microphone gain, I seem to have problem making the silence
> detection work.  
> 
>  
> 
> The speech detection works well for the rare dish sample, which has
> very low noise amplitudes in silence regions.  However, if the
> microphone gain is set to really high, noise samples are taken as
> speech, as indicated by the non-zero return value from speex_encode()
> calls.
> 
>  
> 
> I had VAD turned on.
> 
>  
> 
> Are there other settings that I need to adjust?  If not, could someone
> recommend a suppression solution in this case ?
> 
>  
> 
>  
> 
> Thank you,
> 
>  
> 
>  
> 
> Tongbiao Li
> 
>  
-- 
Jean-Marc Valin, M.Sc.A., ing. jr.
LABORIUS (http://www.gel.usherb.ca/laborius)
Université de Sherbrooke, Québec, Canada
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

> This works really well for white noise reduction.  However 
> what I've noticed was the amplitudes of normal speech samples 
> also get reduced.
Noticed this as well recently.

-- Daniel, Epic Games Inc.

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

Le dim 31/08/2003 à 20:12, Daniel Vogel a écrit :
> > This works really well for white noise reduction.  However 
> > what I've noticed was the amplitudes of normal speech samples 
> > also get reduced.
> 
> Noticed this as well recently.
This is probably due to the AGC (Adaptive Gain Control) that's
integrated with the denoiser. I'll try adding an option to control that.

        Jean-Marc


-- 
Jean-Marc Valin, M.Sc.A., ing. jr.
LABORIUS (http://www.gel.usherb.ca/laborius)
Université de Sherbrooke, Québec, Canada


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 190 bytes
Desc: Ceci est une partie de message numériquement signée
Url :
http://lists.xiph.org/pipermail/speex-dev/attachments/20030831/18281086/signature-0001.pgp