LPC filter coefficients

[Andrea C]

I’m not sure if I understood the theory correctly. Do we calculate the filter coefficients as if they were the filter producing the speech we’re trying to synthesize? If we took the already recorded speech as the desired result and calculate the coefficients of the filter as to have <pulse train> -> <filter> -> <recorded speech>, going back through the filter would give us back the pulse train, so what’s the basis on which we calculate the coefficients?

[Simon]

Do we calculate the filter coefficients as if they were the filter producing the speech we’re trying to synthesise?

Yes, that’s correct – in effect, we fit the filter to the spectral envelope of the speech.

Would inverse filtering the recorded speech result in a pulse train as the excitation signal?

No. The filter is simple and cannot model speech perfectly. The error in this modelling is captured in the residual signal. The residual is a waveform. For voiced speech, it will be more similar to a pulse train than the speech was, but not exactly a pulse train.

[Author]

Posts