I am currently a bit confused. I am currently trying to understand the MFCC extraction process, and don't understand how the Mel filter banks only provide a scalar output from each filter bank. What does this scalar value mean? I know the filters aren't infinitely narrow, thereby only allowing one frequency so what about the frequencies around the center frequency?
Scalar Value from the output of each mel filter bank is accumulated scaled magnitudes of frequencies present in mel filters range/bandwidth.