D. H. Whalen

The universality of intrinsic F0 of vowels

Abstract The tendency for high vowels such as [i] and [u] to have higher fundamental frequencies (F0s) than low vowels such as [a] has been found in every language so far in which it has been sought. This includes 31 languages representing 11 of the worlds 29 major language families (as defined by Crystal, 1987 ). While the size of the intrinsic F0 (IF0) effect varies from study to study, the differences seem to derive from differences in the study design, especially in the number of subjects. The effect appears larger for female speakers when expressed in Hz, but it is, instead, larger for males when the results are expressed in semitones. The size of the languages vowel inventory did not significantly affect the size of IF0. One other universal, though, is that the effect disappears at the low end of a speakers F0 range. The consistency of intrinsic F0 across languages argues that the effect is truly intrinsic; that is, it is not a deliberate enhancement of the signal but rather a consequence of successfully forming a vowel.

Information for Mandarin tones in the amplitude contour and in brief segments

While the tones of Mandarin are conveyed mainly by the F0 contour, they also differ consistently in duration and in amplitude contour. The contribution of these factors was examined by using signal-correlated noise stimuli, in which natural speech is manipulated so that it has no F0 or formant structure but retains its original amplitude contour and duration. Tones 2, 3 and 4 were perceptible from just the amplitude contour, even when duration was not also a cue. In two further experiments, the location of the critical information for the tones during the course of the syllable was examined by extracting small segments from each part of the original syllable. Tones 2 and 3 were often confused with each other, and segments which did not have much F0 change were most often heard as Tone 1. There were, though, also cases in which a low, unchanging pitch was heard as Tone 3, indicating a partial effect of register even in Mandarin. F0 was positively correlated with amplitude, even when both were computed on a pitch period basis. Taken together, the results show that Mandarin tones are realized in more than just the F0 pattern, that amplitude contours can be used by listeners as cues for tone identification, and that not every portion of the F0 pattern unambiguously indicates the original tone.

Audiovisual Processing in Children with and without Autism Spectrum Disorders

Fifteen children with autism spectrum disorders (ASD) and twenty-one children without ASD completed six perceptual tasks designed to characterize the nature of the audiovisual processing difficulties experienced by children with ASD. Children with ASD scored significantly lower than children without ASD on audiovisual tasks involving human faces and voices, but scored similarly to children without ASD on audiovisual tasks involving nonhuman stimuli (bouncing balls). Results suggest that children with ASD may use visual information for speech differently from children without ASD. Exploratory results support an inverse association between audiovisual speech processing capacities and social impairment in children with ASD.

Subcategorical phonetic mismatches slow phonetic judgments

When an [s] or an [š] fricative noise is combined with vocalic formant transitions appropriate to a different fricative, the resulting consonantal percept is usually that of the noise. To see if the mismatch affects processing time, five experiments were run. Three experiments examined reaction time for identification of [s] and [š], as well as the whole syllable (in one experiment) or only the vowel (in the others). The stimuli contained either appropriate or inappropriate formant transitions, and the vowel information in the noise was either appropriate or not. Subjects were significantly slower in all tasks in identifying stimuli with inappropriate transitions or inappropriate vowel information. Similar results were obtained with stop-vowel syllables in which the release bursts of syllable initials [p] and [k]were transposed in syllables containing the vowels [a] and [u], In the fifth experiment, enough silence was introduced between the initial fricatives and vocalic segment for the vocalic formant transitions to be perceived as a stop (e.g., [stu] from [su]). Mismatched transitions still had an effect on reaction time, as did mismatches of vowel quality. The results indicate that listeners take into account all available cues, even when the phonetic judgment seems to be based on only some of the cues.

Intonational differences between the reduplicative babbling of French- and English-learning infants

The two- and three-syllable reduplicative babbling of five French-learning and five English-learning infants (0;5 to 1;1) was examined in two ways for intonational differences. The first measure was a categorization into one of five categories (RISING, FALLING, RISE-FALL, FALL-RISE, LEVEL) by expert listeners. The second was the fundamental frequency (F0) from the early, middle and late portion of each syllable. Both measures showed significant differences between the two language groups. 65% of the utterances from both groups were classified as either rising of falling. For the French children, these were divided equally into the rising and the falling categories, while 75% of those utterances for the English children were judged to have falling intonation. Proportions of the other three categories were not significantly different by language environment. In both languages, though, three-syllable utterances were more likely to have a complex contour than two-syllable ones. Analysis of the F0 patterns confirmed the perceptual assessment. Several aspects of the target languages help explain these intonational differences in prelinguistic babbling.

Effects of vocalic formant transitions and vowel quality on the English [s]–[š] boundary

The effects of the vocalic portion of fricative-vowel syllables on the perception of alveolar and palatal fricatives were examined. The fricative noises were synthesized to represent a continuum from [s] to [ŝ]; the vowels ranged from [u] to [i] through [1] and [ü]. The vocalic formant transitions were of two types, those appropriate to [s] and those to [ŝ]. All stimuli were presented in forced-choice labeling tests. The boundary between [s] and [ŝ] for English-speaking listeners was found to vary as a function both of transitions and of vowel. The effect of the transitions was clear and straightforward: An ambiguous fricative noise was heard more often as [s] before [s] transitions, and as [ŝ] before [ŝ] transitions. The quality of the vowel clearly had an effect, but the interpretation of the effect in terms of the perception of coarticulation was not clear. The responses of listeners who were unfamiliar with languages which use [ü] and/or [1] distinctively were not significantly different from those of listeners who were familiar with such languages.

Subcategorical phonetic mismatches and lexical access

The place of phonetic analysis in the perception of words is unclear. While some theories assume fully specified phonemic strings as input, other theories assume that little analysis occurs. An earlier experiment by Streeter and Nigro (1979) produced evidence, based on auditorily presented words with misleading acoustic cues, that lexical decisions were based on mostly unanalyzed patterns, since word judgments were delayed by misleading information whereas non word judgments were not. The present studies expand that work to a different set of cues, and to cases in which the overriding cue came first. An additional task, auditory naming, was used to examine the effects when the decision stage is less demanding. For the lexical decision task, misleading information slowed the responses, for both words and nonwords. In the auditory naming task, only the slower responses were affected. These results suggest that phonetic conflicts are resolved prior to lexical access.

Perception of pitch location within a speaker’s F0 range

Fundamental frequency (F0) is used for many purposes in speech, but its linguistic significance is based on its relation to the speakers range, not its absolute value. While it may be that listeners can gauge a specific pitch relative to a speakers range by recognizing it from experience, whether they can do the same for an unfamiliar voice is an open question. The present experiment explored that question. Twenty native speakers of English (10 male, 10 female) produced the vowel /a/ with a spoken (not sung) voice quality at varying pitches within their own ranges. Listeners then judged, without familiarization or context, where each isolated F0 lay within each speakers range. Correlations were high both for the entire range (0.721) and for the range minus the extremes (0.609). Correlations were somewhat higher when the F0s were related to the range of all the speakers, either separated by sex (0.830) or pooled (0.848), but several factors discussed here may help account for this pattern. Regardless, the present data provide strong support for the hypothesis that listeners are able to locate an F0 reliably within a range without external context or prior exposure to a speakers voice.

Vowel and consonant judgments are not independent when cued by the same information

Despite many attempts to define the major unit of speech perception, none has been generally accepted. In a unique study, Mermelstein (1978) claimed that consonants and vowels are the appropriate units because a single piece of information (duration, in this case) can be used for one distinction without affecting the other. In a replication, this apparent independence was found, instead, to reflect a lack of statistical power: The vowel and consonant judgments did interact. In another experiment, interdependence of two phonetic judgments was found in responses based on the fricative noise and the vocalic formants of a fricative-vowel syllable. These results show that each judgment made on speech signals must take into account other judgments that compete for information in the same signal. An account is proposed that takes segments as the primary units, with syllables imposing constraints on the shape they may take.

Can Children With Autism Spectrum Disorders “Hear” a Speaking Face?

This study used eye-tracking methodology to assess audiovisual speech perception in 26 children ranging in age from 5 to 15 years, half with autism spectrum disorders (ASD) and half with typical development. Given the characteristic reduction in gaze to the faces of others in children with ASD, it was hypothesized that they would show reduced influence of visual information on heard speech. Responses were compared on a set of auditory, visual, and audiovisual speech perception tasks. Even when fixated on the face of the speaker, children with ASD were less visually influenced than typical development controls. This indicates fundamental differences in the processing of audiovisual speech in children with ASD, which may contribute to their language and communication impairments.