Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It only takes a minute to sign up.
Sign up to join this community
In the context of speech recognition (recognizing individual speech sounds), the pitch of a certain person can change at different times.
Excerpt from Statistical Signal Processing by Steven Kay:
This is a natural variability due to the nature of human speech. The spectral envelope will not change with pitch since the Fourier transform of a periodic signal is a sampled version of the Fourier transform one one period of the signal.
What does the part in bold above mean?
The spectral envelope, which determines where the formant frequencies are, is determined by the shape of the mouth, tongue, lips, and nasal coupling. That is independent of the pitch, which is dependent on the tension or stiffness of the vocal cords.
