Correlation of MFCCs

[Nicole M]

Is it true that mel-frequency cepstral coefficients do not correlate with each other or is there just *less* correlation than when using raw filterbank magnitudes and we assume that the correlation is zero?

[Simon]

There is just less correlation than between filterbank outputs. So much less that we assume there is none (or at least, none worth modelling)!

In reality, there is of course some remaining correlation. In an advanced Automatic Speech Recognition course we would look at ways to model covariance without having a full covariance matrix for each and every Gaussian (because that would be too many parameters). We might share covariances between Guassians, or do clever things with the diagonal vs. off-diagonal entries. But all of that is well beyond the scope of the Speech Processing course.

[Author]

Posts