What if i apply high pass & low pass filter to audio file?

Hi,

I want an exaplantion about what if i apply low pass butterworth 5th order filter to an audio file? What happens if I apply high pass butterworth 5th order filter to an audio file? And lastly what happens if I apply white noise to an audio file?

I don't want the code. I want to understand the concept first so can anyone please send me good explaination about this?

Answer 1

[Star Strider]

A low-pass filter would delete the frequencies above the filter cutoff.

A high-pass filter would delete the frequencies below the filter cutoff.

(You can cascade the filters if you wanted to do both, but it would be more efficient to use a bandpass filter in that instance.)

Adding white noise would add noise to the signal evenly across the spectrum.

[Rajith]

I got to know this basic and can you elabarate? Why I should apply these filter before extarction of an audio file?

[Star Strider]

I’m not certain I understand what you mean by ‘Why I should apply these filter before extarction of an audio file?’

You have to load the file into your MATLAB workspace in order to filter it. It also has to be of the double data type in order to filter it (I can’t find documentation that says any of the filter functions support other data types).

[Rajith]

Ya you are absoultely correct. I want to extract MFCC and PLP features for the audio file before extracting I have beed told to apply low pass, high pass filter and white noise. Which I don't get the extact meaning?

[Star Strider]

I’m not familiar with MFCC and PLP features, but Dan Ellis at Columbia (NYC, U.S.) obviously is. His page PLP and RASTA (and MFCC, and inversion) in Matlab using melfcc.m and invmelfcc.m seems a good place to start. (Please cite his page if you use his routines.)

It is probably best for you to search the Internet to find information most relevant to your particular applications. Feature extraction from audio signals is not an area of my expertise, but it looks interesting.