Quantification of glottal and voiced speech harmonics- to-noise ratios using cepstral-based estimation

Cepstral analysis is used to estimate the harmonics-to-noise ratio (HNR) in speech signals. The inverse Fourier transformed liftered cepstrum approximates a noise baseline from which the harmonics-to-noise ratio is estimated. The present study highlights the manner in which the cepstrum-based noise baseline estimate is obtained, essentially behaving like a moving average filter applied to the power spectrum for voiced speech. As such, the noise baseline, which is taken to approximate the noise excited vocal tract, is also shown to be influenced by the window length and the shape of the glottal source spectrum. Two existing estimation techniques are tested systematically for the first time using synthetically generated glottal flow and voiced speech signals, with a priori knowledge of the HNR. The source influence is removed using preemphasis to obtain an improved noise baseline fit. The results indicate accurate HNR estimation using the new approach.