Perception Of Acoustic Cues Research Articles

Acoustic cues in perception of reduced speech

In conversational speech, sounds undergo reduction and deletion, which challenges listeners to decode reduced forms using various cues. Previous research argues for the predominance of acoustic cues in a study of listeners’ perception of reduced speech such as “he’s” vs. “he was”, which can both be realized [ɨz]. Analyzing data from a previous perception experiment (Warner et al., Brain Sci., 2022), this study explores what types of acoustic cues listeners use to process reduced speech. Ambiguous [ɨz] with low second formant (F2) or long duration might be perceived as past due to an additional consonant (/w/) and potential contractions (low F2 as a cue to /w/). Low F2 is expected to result in smaller Bark F2-F1 and larger Bark F3-F2. Current study tests whether the stimuli show acoustic differences in the predicted direction and whether the formant and duration measures correlate with listeners’ identification of tense. It was found that there are no significant acoustic differences between past and present forms due to reduction, but that listeners significantly associated smaller Bark F2-F1 with past tense in singular verbs and longer duration with past tense in plural verbs. This confirms listeners' use of the predicted acoustic cues in perceiving reduced speech.

The impact of one's own voice and production skills on word recognition in a second language.

Second language (L2) learners often speak with a strong accent, which can make them difficult to understand. However, familiarity with an accent enhances intelligibility. We propose that L2 learners are even more familiar with their own accented speech patterns and may thus understand self-produced L2 words better than others' accented productions, presumably because of adaptation. This hypothesis was tested by asking German learners of English to identify English words from minimal pairs that are distinguished by difficult L2 sound contrasts. Words had been spoken by the learners themselves or other learners who produced the contrasts equally well. Self-produced words were identified significantly better than others' productions. A second experiment revealed that better producers can exploit acoustic cues in perception more than poor producers, especially when the produced acoustic cues to the minimal pairs were clearly differentiated. The self-benefit, however, did not depend on production skills. We conclude that L2 learners adapt not only to their L1 accent in general but also to their own specific speech patterns. Speculating about L2 acquisition more generally, these results may raise the question whether adaptation to own, accented productions may be one reason why learners have difficulties to improve their pronunciation, since they may not notice a need to improve. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Acoustic Cues in Korean High School Students’ L2 Phoneme Perception and Speech Segmentation

This study investigates perceptual patterns in the categorization of English stop consonants regarding VOT and F0 by Korean learners of English. The effect of perception patterns on the segmentation of English sound sequences was also examined. With the stimuli of English stop consonants followed by conflicting vocalic cues, the subjects showed a degree of vowel?dependent perception in certain types of combinations, ignoring VOT cues. Also, students with high vowel?dependency performed significantly worse than those with low vowel?dependency on the word boundary perception test regarding VOT cue, implying that the native cue weighting strategy hinders L2 speech segmentation. The findings of this study show that the transfer of native cue weighting strategy may impede the use of crucial acoustic cues in L2 segmentation. Therefore, instruction based on the use of acoustic cues in perception is needed rather than instruction based on phonemes.

Ambiguity related to French liaisons: The role of bottom-up and top-down processes

French liaison is a phonological process involving the surfacing of an underlying floating consonant as the onset of a vowel-initial word when the word is preceded by a liaison-causing word. We used vowel- and consonant-initial ambiguous targets following four liaison-related contexts /z/, /n/, /t/, /r/ (e.g., ces onches - ces zonches). Targets included nouns and pseudo-nouns. Quebec-French-speaking adults performed two tasks (production, discrimination). One bottom-up (acoustic cues) and three top-down (noun token frequency, word onset probability, contextual liaison knowledge) factors were investigated in the production task. Participants had to produce the last word upon hearing each phrase. Their productions thus reflected their interpretation of the onset of the liaison-ambiguous words. Perception of acoustic cues was also tested in the discrimination task: participants judged if two phrases were different or same. No perception of any acoustical distinction was revealed in the tasks: participants often produced the intended target incorrectly, and differently intended phrasal pairs were judged same (e.g., ces onches - ces zonches). Effects of top-down information were found in the production task. Among the top-down factors, liaison knowledge related to liaison-causing words had a dominant impact on participants’ interpretation of the ambiguous phrases, showing the importance of contextual knowledge in lexical recognition.

Perception of acoustic cues in contrastive focus.

This research examines the ability of native English speakers to identify meanings associated with contrastive focus. Two conditions are compared: adjective and noun focus. Subjects hear either “I want the RED shirt” or “I want the red SHIRT,” where the focused element receives an L+H* pitch accent and a longer duration. They must then choose which of two continuations best follows the auditory stimulus (e.g., “I don’t like the blue shirt” versus “I don’t like the red pants”). Preliminary results, based on 23 subjects listening to 10 sentences each, show that despite substantial acoustic differences between the focused and nonfocused words for both the adjective and noun conditions, each condition’s result is vastly different. As a group, they excel at identifying the adjective‐focused condition, but do poorly in the noun‐focused condition. Closer examination reveals a bimodal distribution on the noun condition: One group has a high success rate (between 80%–100%) and another has a low success rate (between 20%–40%). For the low success group, conflicting acoustic cues may result in a misinterpretation as phrase‐final lengthening and/or natural focus on the object, while these cues are being overridden in the high success group.

Recognition of emotion in Japanese, Western, and Hindustani music by Japanese listeners1

Abstract: Japanese listeners rated the expression of joy, anger and sadness in Japanese, Western, and Hindustani music. Excerpts were also rated for tempo, loudness, and complexity. Listeners were sensitive to the intended emotion in music from all three cultures, and judgments of emotion were related to judgments of acoustic cues. High ratings of joy were associated with music judged to be fast in tempo and melodically simple. High ratings of sadness were associated with music judged to be slow in tempo and melodically complex. High ratings of anger were associated with music judged to be louder and more complex. The findings suggest that listeners are sensitive to emotion in familiar and unfamiliar music, and this sensitivity is associated with the perception of acoustic cues that transcend cultural boundaries.

Development and testing of courseware in acoustic phonetics

Sensimetrics Corporation has been engaged in developing and testing interactive, multimedia educational software in acoustic phonetics for more than three years. The course software contains a nonmathematical presentation of major topics in acoustic phonetics, a lab in which students can record and analyze their own speech, and a library in which students can look up technical terms, hear and see spectrograms of IPA examples, and find readings related to the topics covered in the course software. Topics covered in the software include reading spectrograms, basic vowel acoustics, articulatory-acoustic relations in vowels, speaker differences in vowels, acoustic characteristics of constants, aerodynamic effects in consonant production, perception of acoustic cues, identification and discrimination experiments, native and nonnative perception of consonants, and perceptual normalization of vowels. Faculty volunteers at various educational institutions used components of the course and solicited anonymous, written feedback from over 200 graduate and undergraduate students in Communication Disorders/Speech and Hearing Science during the ’94–’95 and ’95–’96 academic years. The volunteers also provided written comments on the course software. The results indicate that the course software is a useful adjunct to traditional teaching methods. [Work supported by NIH Grant No. 9 R44 MH 51970-02.]

Perception of /d,n,l/ by impaired‐ and normal‐hearing listeners

Eighteen moderately to profoundly hearing‐impaired and 11 normal‐hearing listeners were studied for their use of acoustic cues for perception of /d,n,l/ in /aeCae/ tokens extracted from spoken sentences. The cues were in segments associated with the /d,n,l/ occlusions and with the transitions of the adjacent vowels. The segments' contribution to /d,n,l/ distinctions were examined from identification tests of the /aeCae/ tokens with consonant occlusion and transition segments degraded singly or in combination. Specifically, the consonant segments were deleted or replaced by a synthetic neutral segment; the transition segments were neutralized via substitution of adjacent pitch periods from the respective vowels. The results with normal‐hearing listeners showed that the presence of any one of the consonant or transition segments supported a moderate to high level of correct /d,n,l/ identification, depending on the phoneme. In contrast, the hearing‐impaired listeners showed subnormal identification of intervocal...

Developmental aspects of the perception of acoustic cues in determining the voicing feature of final stop consonants.

The developmental use of vowel duration, final transition and voicing during closure as cues to voicing in final stop consonants were investigated, using 10 three-year-old, 10 six-year-old and 10 adult subjects. The stimuli were alterations of eight stop-vowel-stop words. The presentation of each stimulus item was response contingent. The resulting data supported the ability of adults and children to use both temporal and spectral cues to acoustic/ phonetic distinctions. However, three-year olds relied more on temporal cues, six-year olds relied more on spectral cues, while adults used both spectral and temporal cues in judging the voicing feature of final stop consonants.

Acoustic Phonetics: A Course of Basic Readings

D. B. Fry has edited a basic course of readings on the acoustics of speech. The collection includes all the important classical papers in the field. It is carefully structured to present the student with a coherent picture of the relations between language units and the corresponding sound-waves and to explain the laws that govern these relations. He includes extracts which explain the generation of sound-waves by the speech-mechanism, the methods of acoustic analysis of speech, and the operation of the sound spectograph (with excerpts from the first published accounts of the instrument). The volume also illustrates the contribution to the general study of language made by research on speech perception. There are accounts of speech synthesis, and of experiments on rhythm, intonation and the perception of acoustic cues.

Duration of oral port constriction in normal and cleft palate speech.

Duration of constriction, determined from measurements of intraoral pressure, was studied in 25 cleft palate and 20 normal subjects. The cleft palate group consisted of 15 subjects who achieved adequate velopharyngeal closure and 10 subjects who did not. All subjects were able to produce the test consonants intelligibily. Results reveal that the duration of oral port constriction is longest in cleft palate speakers with adequate closure and shortest in normals. It is possible that increasing the duration of constriction may reinforce sound generation and thereby improve consonant perception. The shorter intervals of constriction observed in the inadequate closure group may be due to the increase in nasal emission which presumably results when oral port constriction is prolonged. Within each palate group voiceless fricative consonants exhibited the longest intervals of oral port constriction. This is presumed to be due to the need for sufficient airflow through the oral port in order to provide the necessary acoustic cues for perception of these voiceless sounds.