I am working on speech enhancement and I want to compare the phase and magnitude between two signals. I compute the STFT for both, and comparing the magnitudes is easy via mean-squared error or some other metric. However, for the phase, there seems to be no precedent in the literature for how to check if the two phases are close or distant. I have looked into it quite a bit but there doesn't seem to be a consensus on how to measure the distance between phases in this context. Any help would be greatly appreciated.

  • $\begingroup$ What's the purpose of the phase comparison ? The human air is very insensitive to monaural phase: two signals can have completely different phases and still sound almost identical . What are you trying to learn by comparing phases ? $\endgroup$
    – Hilmar
    Mar 2, 2023 at 23:08


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.