No abstract

Relatively few studies have directly examined the use of vowel-inherent spectral change by second-language learners, perhaps because it represents one of the subtler cues to vowel identity. Nevertheless, understanding non-native listeners’ perception of this cue and its integration with other cues to vowel identity can be regarded as a method of investigating the mastery of cues to vowel perception that is needed for native-like perception. While several studies of second-language speech perception have demonstrated differences in cue use by second-language learners, our investigation of the use of dynamic spectral cues and temporal cues by native and non-native listeners suggests that relatively early learners of English as a second language do not appear to use the dynamic spectral cue differently from native English speakers when other cues are preserved. Instead, early learners’ vowel perception appears to be less robust to removal of multiple cues. This apparent difference in even early learners’ ability to use cues to vowel identity independently of one another or to change listening strategy when one cue is degraded may explain a portion of the increased challenge that even relatively early learners of a second language appear to experience in understanding speech in noisy environments. [Work supported by NIH.]

Vowel-inherent spectral change (VISC) refers to vowel-intrinsic formant movement across a vowel steady state. VISC has been shown to (1) be consistent across talkers within a given dialect, (2) vary regularly across vowels within a dialect, (3) vary regularly across dialects, and (4) be necessary for peak vowel-identification accuracy. Hence, VISC has become accepted as a phonetic feature of monophthong vowels of North American English. VISC is typically portrayed using averages across tokens and talkers, highlighting regularity but potentially masking individual differences. To understand vowel production by second-language learners, we were particularly interested in such individual variation. In analyzing individual differences for neighboring target vowels, we found no single time point at which all sets of target vowel tokens were well distinguished from one another. However, looking across three time points, all native-speaker vowel sets were well distinguished from each possible neighbor set at some time point. Thus, VISC can be seen as the steps in a sort of dance, as each vowel moves to avoid overlapping with another, ultimately causing overlap with another and then more movement. This perspective is compatible with models of efficient coding and stochastic and/or exemplar based models of speech production and perception. [NIH-NIDCD #1R03DC005561-01A1.]

The degree to which vowel duration and time-varying spectral change cues are used in American English vowel identification was investigated for three groups of listeners: (1) monolingual American English listeners; (2) proficient Spanish-English bilinguals; and (3) less proficient Spanish-English bilinguals. Digital manipulation and high-fidelity synthesis procedures (STRAIGHT) were used to create six versions of six target items (bead, bid, bayed, bed, bad, and bod). These consisted of the unaltered whole word and five versions of the isolated vowel: (1) natural vowel—unaltered; (2) synthetic vowel with no cue alteration; (3) synthetic vowel with neutral duration; (4) synthetic vowel with flattened formants; and (5) synthetic vowel with neutral duration and flattened formants. Isolated-vowel stimuli were presented to listeners for identification in two 120-trial blocks, followed by 48 whole-word stimuli. Significant between-group differences in percent-correct performance and a significant group by listening condition interaction were found, suggesting that cue weighting may vary across groups and that some bilinguals’ phoneme representations may be less robust than those of monolinguals. Analysis of confusion matrix data will be used to further explore these hypotheses by examining patterns of confusions across group and condition and extent of individual differences across subjects within a group. [Work supported by NIH-NIDCD 1R03DC005561-01A1.]