How do I remove background noise from a sound wave?

I have a sound wave y(1:441000) gathered using a microphone and I have background n(1:441000) also gathered by the microphone. I have tried removing the background noise using a script something like:

Y=fft(y);

N=fft(n);

Yclean=Y-N;

yClean=ifft(Yclean);

However, yClean is not correct and is backwards in time. Do you have any suggestions?

Thanks,

Dave

Answer 1

[Pedro Villena]

Create and Implement LMS Adaptive Filter to remove the filtered noise from desired signal

mtlb_noisy = y;
noise = n;

% Define Adaptive Filter Parameters
filterLength = 32;
weights = zeros(1,filterLength);
step_size = 0.004;

% Initialize Filter's Operational inputs
output = zeros(1,length(mtlb_noisy));
err = zeros(1,length(mtlb_noisy));
input = zeros(1,filterLength);

% For Loop to run through the data and filter out noise
for n = 1: length(mtlb_noisy),

      %Get input vector to filter
      for k= 1:filterLength
          if ((n-k)>0)
              input(k) = noise(n-k+1);
          end
      end

      output(n) = weights * input';  %Output of Adaptive Filter
      err(n)  = mtlb_noisy(n) - output(n); %Error Computation
      weights = weights + step_size * err(n) * input; %Weights Updating 
  end

yClean = err;

Introduction to Simulink for Signal Processing

[Tahira Batool]

And what if one does not have a separate noisy signal to be removed from an original signal ,then how can we remove background noise from a signal?

Answer 2

[Umair Nadeem]

It would be easier if you could upload the noisy signal too. Save the variable y which supposedly has the noisy signal in a .mat file using save command and attach it with your post. Some frequency analysis could be done if the signal is available.

Also try to provide info about the signal frequency (if known), and the sampling frequency which you used to sample the data.

Answer 3

[pinreddy chaitanya]

 weights = weights + step_size * err(n) * input; %Weights Updating

what is the use of this line

[Albin Lindmark]

It is to update the weights of the adaptive filter.

Answer 4

[pravin m]

mtlb_noisy = y;

noise = n;

% Define Adaptive Filter Parameters

filterLength = 32;

weights = zeros(1,filterLength);

step_size = 0.004;

% Initialize Filter's Operational inputs

output = zeros(1,length(mtlb_noisy));

err = zeros(1,length(mtlb_noisy));

input = zeros(1,filterLength);

% For Loop to run through the data and filter out noise

for n = 1: length(mtlb_noisy),

%Get input vector to filter

for k= 1:filterLength

if ((n-k)>0)

input(k) = noise(n-k+1);

end

end

output(n) = weights * input'; %Output of Adaptive Filter

err(n) = mtlb_noisy(n) - output(n); %Error Computation

weights = weights + step_size * err(n) * input; %Weights Updating

end

yClean = err;