0
$\begingroup$

The Burg algorithm implemented in PRAAT for formant estimation also works via LPC. First the sound is resampled to a sampling frequency of twice the value of maximum formant. Then a pre-emphasis is applied and finally, PRAAT applies a Gaussian like window to compute linear prediction coefficients through the Burg algorithm.

I understand that pre-emphasis is helpful for formant extraction by LPC since the human voice is not spectrally flat, but am confused about why Praat is resampling.

https://github.com/praat/praat/blob/bac77ad76481c65703ade8d7ddb190e9d67048d6/fon/Sound_to_Formant.cpp#L381

$\endgroup$
0
$\begingroup$

You want to resample down to alias frequencies higher than your maximum desired formant.

If you understand aliasing in the z-transform and the transfer function of a system then this should be obvious.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.