Perceptual interaction of the harmonic source and noise in voice

Abstract

Although the amount of inharmonic energy (noise) present in a human voice is an important determinant of vocal quality, little is known about the perceptual interaction between harmonic and inharmonic aspects of the voice source. This paper reports three experiments investigating this issue. Results indicate that perception of the harmonic slope and of noise levels are both influenced by complex interactions between the spectral shape and relative levels of harmonic and noise energy in the voice source. Just-noticeable differences (JNDs) for the noise-to-harmonics ratio (NHR) varied significantly with the NHR and harmonic spectral slope, but NHR had no effect on JNDs for NHR when harmonic slopes were steepest, and harmonic slope had no effect when NHRs were highest. Perception of changes in the harmonic source slope depended on NHR and on the harmonic source slope: JNDs increased when spectra rolled off steeply, with this effect in turn depending on NHR. Finally, all effects were modulated by the shape of the noise spectrum. It thus appears that, beyond masking, understanding perception of individual parameters requires knowledge of the acoustic context in which they function, consistent with the view that voices are integral patterns that resist decomposition.

Figure 1

(Color online) Manipulations of the harmonic voice source spectrum. Listeners adjust the slope of H2–Hn by typing the desired slope value into a box (not shown) and then clicking the point labeled with a double arrow. Note that H1–H2 remains constant throughout manipulations of H2–Hn.

Figure 2

Variations in sensitivity to changes in NHR and H2–Hn as a function of baseline H2–Hn. The y axis shows the ratio of the JND to range for the NHR (an index of listeners’ overall sensitivity), and the x axis shows baseline H2–Hn. (a) Sensitivity to changes in NHR. (b) Sensitivity to changes in H2–Hn. Values for NHR = −40 dB (noise free) are plotted with filled circles; open squares represent values when the NHR = −30 dB; asterisks show values when the NHR = −20 dB; and filled triangles indicate values when the NHR = −10 dB (very noisy). Ellipses enclose points that do not differ significantly. See text for fuller description.

Figure 3

Representative noise spectra. Units on the y axis are arbitrary. (a) A typical falling spectrum. (b) A typical rising spectrum. (c) A typical flat spectrum.

Figure 4

(Color online) Changes in discrimination accuracy (measured by d′) for the three noise sources, as a function of changes in (a) NHR and (b) H2–Hn. See text for more discussion.