On the dissociation of spectral and temporal cues to the voicing distinction in initial stop consonants

Abstract

A rising first-formant (F1) transition may be an important cue to the voiced-voiceless distinction for syllable-initial, prestressed stop consonants in English. Lisker showed that the acoustic manipulations suggesting a role for F1 have involved covariation of the onset frequency of F1 with the duration, and hence the frequency extent, of the F1 transition; he argued that effects hitherto ascribed to the transition are more properly attributed to its onset. Two experiments [using human subjects] were reported in which F1 onset frequency and transition duration/extent were manipulated independently. The results confirmed Lisker''s suggestion that the major effect of F1 in initial voicing contrasts was determined by its perceived frequency at the onset of voicing, and showed that a periodically excited F1 transition was not, per se, a positive cue to voicing. The relative levels and the frequencies at onset of voicing F1 and F2 were manipulated. The influences on the perception of stop-consonant voicing that resulted were determined specifically by the frequency of F1 not by its absolute or relative level or the overall distribution of energy in the spectrum. The results demonstrated a complementary relationship between perceptual cue sensitivity and production constraints; in production the VOT characterizing a particular stop consonant varied inversely with the degree of vocal-tract constriction, and hence with the frequency of F1, required by the phoneme following the stop; in perception the lower the frequency of F1 at the onset of voicing, the longer the VOT that is required to cue voicelessness. In this way the inclusion of F1 onset frequency in the cue repertoire for voicing and the establishment of the cue trading relationship reduced the problem of contextual variation that would be met were VOT alone, or some other amalgam of cues, the only basis of the voicing distinction.