Rules | Recent posts | topic RSS | Search | Register  | Log in

Spectral Subtraction

 
Post new topic  Reply to topic    EDAboard.com Forum Index -> Digital Signal Processing
Author Message
ednan



Joined: 13 Feb 2007
Posts: 131


Post02 Aug 2008 17:05   Spectral Subtraction

Hello to all members....

i need help in ..Speech Enhancement Using Spectral Subtraction method

my question is ...
(i) What are steps using to do Spectral Subtraction Method ......

(ii) if we use Function for Spectral Subtraction
[OUT,PO]=SS(IN,FS,P)

Please tell me wat is P and PO...
please if ani 1 know about this ...plz help me...
Regards
Back to top
Eugen_E



Joined: 29 Nov 2004
Posts: 329
Helped: 30


Post02 Aug 2008 23:10   Re: Spectral Subtraction

You're talking about Matlab?

Spectral subtraction means:
- obtaining the steady spectrum, without useful signal present - this can be made by long time spectrum averaging;
- subtracting the steady spectrum from the instantaneous spectrum.
Thus the noise floor and other continous unusefull signals are eliminated.
Back to top
ednan



Joined: 13 Feb 2007
Posts: 131


Post03 Aug 2008 9:29   Re: Spectral Subtraction

Yes m Taking about MATLAB ...i Cant Undersyand that Code ...


%SPECTRALSUBTRACTION performs speech enhancement using spectral subtraction [OUT,PO]=SS(IN,FS,P)
%
% implementation of spectral subtraction algorithm
% noisy speech IN with fs=FS enhanced by SS and returned in OUT, P is the algorithm parameters as bellow:
% algorithm parameters: t* in seconds, f* in Hz, k* dimensionless
% 1: tg = smoothing time constant for signal power estimate (0.04): high=reverberant, low=musical
% 2: ta = smoothing time constant for signal power estimate used in noise estimation (0.1)
% 3: tw = fft window length (will be rounded up to 2^nw samples)
% 4: tm = length of minimum filter (1.5): high=slow response to noise increase, low=distortion
% 5: to = time constant for oversubtraction factor (0.0Cool
% 6: fo = oversubtraction corner frequency (800): high=distortion, low=musical
% 7: km = number of minimisation buffers to use (4): high=waste memory, low=noise modulation
% 8: ks = oversampling constant (4)
% 9: kn = noise estimate compensation (1.5)
% 10:kf = subtraction floor (0.02): high=noisy, low=musical
% 11:ko = oversubtraction scale factor (4): high=distortion, low=musical


Added after 1 minutes:

Please tell me about P which is the Parmater ...and about PO
Back to top
ednan



Joined: 13 Feb 2007
Posts: 131


Post04 Aug 2008 21:19   Re: Spectral Subtraction

Please please help me ...
Back to top
Post new topic  Reply to topic    EDAboard.com Forum Index -> Digital Signal Processing
Page 1 of 1 All times are GMT + 1 Hour


Abuse
Administrator
Moderators
topic RSS 
sitemap