Convolve vector with audio file

Audio file: y = 537146x2 double

function: hx=1x13839 double

I need to convolve an audio file with the transfer function of an impulse response of an FIR filter to remove a range of frequencies, but I'm not sure how when the audio file vector is so much larger than the function vector.

Answer 1

[Swetha Polemoni]

Hi Matthew Strunks

It is my understanding that you want to convolve your audio signal and FIR filter. Since convolution in time domain is equal to multiplication in frequency domain, you can try first converting both into frequency domain and multiply.

For converting audio signal to frequency domain "stft" (short time fourier transform) can be used. Since stft applies fft on windowed audio signal, characteristiics of audio signal is preserved even if the length of audio signal is large.

For coverting your filter into time domain, "fft" can be used.

Feel free to experiment with window size, so that the result is same as you expected.