Audio processing using LMS algorithm

Hello

I would be grateful if you could help me in the following issue:

I have a "Pure Speech" audio file and a "Noisy" audio file which consists of "Pure Speech" and background noise (passed through a microphone).

I need to extract the background noise (Not Speech) from the "Noisy" file using the "Pure Speech" as the reference signal.

I tried different kinds of LMS and RLS algorithm to do that, but all of these algorithms failed to extract the background noise.

It would be very helpful if you could guide me what the problem is.

Answer 1

[Selva Karna]

Split the channel ,then apply weiner filter

[Meysam Jahanshahi]

thank you for your anwer.

the information of the audio files ara as below:

Noisy Audio:

NumChannels: 1

SampleRate: 48000

BitsPerSample: 24

pure Audio:

NumChannels: 1

SampleRate: 48000

BitsPerSample: 16

The only difference is bit depth. I changed the bit depth of "Noisy Audio" to 16 (as pure audio), but it did not resolve the problem.