Vowel Intelligibility Research Articles

Intelligibility of dysphonic speech in auralized classrooms

Voice disorders reduces speech intelligibility. This study evaluated the effect of noise, voice disorders and room acoustics on vowel intelligibility. Twenty-nine college students listened to 11 vowels in /h/-V-/d/ format. The speech was recorded by three adult females with dysphonia and three adult females with normal voice quality. The recordings were convolved with two oral-binaural impulse responses with 0.4 s and 3.1 s of reverberation time. The intelligibility and the listening easiness were significantly higher in quiet condition, when the speakers had normal voice quality and in low reverberated environments, while the response time of the listener was longer in noise condition.

Perception and Acoustic Studies of Vowel Intelligibility in Dysphonic Speech

This study aimed to (1) examine the effect of dysphonia and background noise on the identifiability of vowels, and (2) evaluate the relationship between the degree of vowel intelligibility and vowel-related acoustic measurements. A speech perception experiment was conducted with speech samples collected from 10 adult females with healthy voices and 10 adult females with dysphonia. The speech material was 11 vowels of American English in /h/-vowel-/d/ format. Cafeteria noise was added to these samples at a signal-to-noise ratio of -6 dB. A total of 10 adults with normal hearing participated in a speech perception experiment, in which the vowels were presented with and without the noise. F1 and F2 frequencies of the vowels were measured, and their relationships with the vowel intelligibility were statistically evaluated. The group-level analysis showed that vowel intelligibility was lower in dysphonic speech than normal speech, both in quiet and at signal-to-noise ratio of -6 dB. The intelligibility was higher for the high vowels than the low vowels. In general, the vowel confusion pattern was similar between normal and dysphonic speech. However, data from a speaker with severe diplophonia showed a distinct confusion pattern. Voice quality ratings did not significantly correlate with the vowel intelligibility. There was a significant correlation between F2 and the vowel intelligibility in quiet. A post-hoc acoustic analysis revealed that vowels of the speakers with lower vowel intelligibility had lower acoustic energy above 1 kHz. Dysphonia negatively affects vowel intelligibility. Low vowels were more vulnerable to the effect of dysphonia than high vowels. Among different types of dysphonic voice qualities, diplophonia appears to be particularly detrimental to vowel intelligibility. F2 significantly correlated with vowel intelligibility; however, this result requires a careful interpretation. Given that the acoustic energy above 1 kHz resulted in better intelligibility in noise, a treatment strategy that targets this frequency region may help improve intelligibility in noise. Future studies should examine the generalizability of this finding to different age and gender groups, and intelligibility as a whole.

Intelligibility Enhancement Assessment and Intervention: a single-case experimental design with two multilingual university students

ABSTRACT Speech-language pathologists (SLPs) may be approached by multilingual speakers wishing to improve their intelligibility in English. Intelligibility is an essential element of spoken language proficiency and is particularly important for multilingual international students given their need to express complex ideas in an additional language. Intelligibility Enhancement aims to improve the intelligibility and acceptability of consonants, vowels and prosody with multilingual speakers who are learning to speak English. This study aimed to describe the Intelligibility Enhancement Assessment and Intervention Protocols and determine whether the intervention changed multilingual university students’ English intelligibility. A multiple-baseline single-case experimental design was applied with direct inter-subject replication across two female participants whose home languages were Vietnamese and Putonghua (Mandarin). English intelligibility was assessed at multiple intervals pre, post and during intervention. The intervention protocol consisted of 11 weekly 1-h sessions with an SLP targeting English consonants, vowels and prosody. Following intervention, both participants displayed increased performance across most measures. For example, the Vietnamese participant’s percentage of consonants correct (PCC) increased from 62.5% to 85.0% in probe keywords. Effect sizes, when comparing baseline and withdrawal phases, were 5.5 for PCC, 4.6 for final consonants, 2.3 for consonant clusters and 1.6 for syllables indicating improvements in all variables measured. Her speech rate reduced, word stress increased in accuracy and she perceived less difficulty communicating in English. These promising results suggest further testing of the Intelligibility Enhancement Protocols is warranted to determine effectiveness as an intervention for multilingual speakers.

What *can* make clear speech clear: Lessons learned from the Ferguson Clear Speech Database

Extensive acoustic and perceptual analyses have been carried out on the materials from the Ferguson Clear Speech Database (FCSD), which was recorded at Indiana University in 2002. The FCSD consists of 41 untrained talkers reading 188 sentences under instructions first to speak in a manner “as much like your normal conversational style as possible” and later to “speak clearly, so that a hearing-impaired person would be able to understand you.” My intent in developing the FCSD was to exploit the expected wide acoustic and perceptual variability among the talkers and use a talker-differences approach to answer the question, “What makes clear speech clear?” In this presentation, I will summarize data from studies of vowel intelligibility, word intelligibility, and perceived sentence clarity along with global and fine-grained acoustic analyses, and discuss how all of these measures are related across the 41 talkers. My hope is that this birds-eye view of the FCSD data will reveal subgroups of talkers in which the talkers adopted certain “profiles” of clear speech acoustic changes that yielded specific helpful perceptual changes. If time permits, I will also review data on perceived talker indexical properties and how they change when talkers speak clearly.

Vowel-Specific Intelligibility and Acoustic Patterns in Individuals With Dysarthria Secondary to Amyotrophic Lateral Sclerosis.

Purpose The purpose of the study was to investigate vowel-specific intelligibility and acoustic patterns of individuals with different severities of dysarthria secondary to amyotrophic lateral sclerosis (ALS). Method Twenty-three individuals with dysarthria secondary to ALS and 22 typically aging individuals participated as speakers. Participants with ALS were divided into 2 severity groups (severe, mild). For vowel-specific intelligibility data, 135 listeners participated in the study. Vowel-specific intelligibility, intrinsic vowel duration, 1st and 2nd formants (F1 and F2), vowel inherent spectral change (VISC), and absolute VISC were examined. Results A significant interaction between severity group and the vowel-specific intelligibility pattern as well as F1, F2 VISC, and absolute F2 VISC was observed. Specifically, individuals with severe dysarthria showed a significantly less intelligible /ɪ/ than /ɛ/, unlike individuals with mild dysarthria and typically aging individuals. In addition, vowel intelligibility of /ɪ/ showed the strongest association to the severity measures in individuals with ALS. A number of vowel-specific findings are reported in the acoustic variables. Acoustic correlates of vowel-specific intelligibility were identified. Conclusion Vowel-specific intelligibility patterns are different across severity groups; particularly, low intelligibility of /ɪ/ was noted in individuals with severe dysarthria. Individuals with dysarthria maintained the acoustic contrast in duration and F1 VISC among vowels but did not maintain the other spectral contrasts. Reduction of acoustic vowel space was observed primarily due to high F1 in high vowels in individuals with severe dysarthria. Regression findings suggest that the high F1 values of high and mid vowels and F2 reduction of high- and mid-front vowels decreased vowel-specific intelligibility. In addition, vowel duration influenced the vowel intelligibility of vowels that required short intrinsic vowel duration. Lastly, F2 VISC influenced the vowel intelligibility of /ɪ/. Overall, the vowel-specific intelligibility pattern is related to both vowel-specific characteristics and group-specific articulatory control dysfunction.

Vowel intelligibility in clear speech produced by electrolaryngeal speakers

This study assessed vowel intelligibility when electrolaryngeal (EL) speakers were instructed to use clear speech (CS). Eighteen consonant-vowel-consonant words containing /i/, /ɪ/, /ɛ/, and /æ/ were spoken by 10 laryngectomees in habitual speech (HS) and CS conditions. A total of 4,320 words across both speech conditions were recorded and then transcribed by 12, naïve listeners. Results indicate that vowel intelligibility was 85.4% (range = 77.8% to 90.6%) when EL speakers used HS compared to 82.7% (range = 75.0% to 92.8%) in CS. This finding suggests that CS does not facilitate improved vowel intelligibility for EL speakers. Future research will seek explanations for the lack of a CS benefit by examining the acoustic changes that occur when EL speakers use CS.

Intelligibility of sinewave consonants

A good deal of experimental work has assessed the intelligibility of sinewave speech (SWS), synthesized by mixing sinusoids that follow the formants of natural utterances. While SWS is clearly intelligible at some level, most SWS work has been conducted using sentences, whose intelligibility is affected by many factors in addition to those related to recognition at the phonetic level. Earlier work [Hillenbrand et al., J. Acoust. Soc. Am, 129, 3991–4000] measuring the intelligibility of SWS vowels in isolated syllables reported an identification rate of 55%, far above chance but ~40 percentage points lower that of the original signals. The present work tested the intelligibility of SWS versions of 23 consonant types in CV and VCV syllables with three vowel types ([ɑ i u]) spoken by one man and one woman. SW signals were generated from unedited envelope peaks rather than formants. Intelligibility averaged across 59 listeners was 59%, with large variability across both listeners (sd = 9.9) and, especially, c...

Vowel intelligibility in children with cochlear implants: An acoustic and articulatory study

In child cochlear implant (CI) users, early implantation generally results in highly intelligible speech. However, for some children developing a high level of speech intelligibility may be problematic. Studies of speech production in CI users have principally been based on perceptual judgment and acoustic measures. Articulatory measures, such as those collected using ultrasound provide the opportunity to more precisely evaluate what makes child CI users more intelligible. This study investigates speech production and intelligibility in children with CI using acoustic and articulatory measures. Ten children with unilateral or bilateral CIs and 13 children with normal hearing (NH) participated in the study. Participants repeated five English vowels (/a/, /e/, /i/, /o/, /u/) with and without auditory feedback. Ultrasound was used to capture tongue positions and acoustic signals were recorded simultaneously. The results showed that, despite quite similar acoustic results, the two speaker groups made different use of the tongue to implement vowel contrasts. Indeed, the tongue position was lower in the feedback OFF condition than the feedback ON condition for all participants, but the magnitude of this difference was larger for CI users than for their NH peers. This difference led to diminished intelligibility scores for CI users. This study shows the limitation of acoustic measurements alone and demonstrates how the use of articulatory measurements can explain intelligibility patterns. Moreover, our results show that when cochlear implantation occurs early in life and auditory feedback is available, CI users' intelligibility is comparable to that of their NH peers.

A Study of the Intelligibility of Japanese Learners' English Vowels

A Study of the Intelligibility of Japanese Learners' English Vowels

The Impact of Contrastive Stress on Vowel Acoustics and Intelligibility in Dysarthria.

To compare vowel acoustics and intelligibility in words produced with and without contrastive stress by speakers with spastic (mixed-spastic) dysarthria secondary to cerebral palsy (DYSCP) and healthy controls (HCs). Fifteen participants (9 men, 6 women; age M = 42 years) with DYSCP and 15 HCs (9 men, 6 women; age M = 36 years) produced sentences containing target words with and without contrastive stress. Forty-five healthy listeners (age M = 25 years) completed a vowel identification task of DYSCP productions. Vowel acoustics were compared across stress conditions and groups using 1st (F1) and 2nd (F2) formant measures. Perceptual intelligibility was compared across stress conditions and dysarthria severity. F1 and F2 significantly increased in stressed words for both groups, although the degree of change differed. Mean Euclidian distance between vowels also increased with stress. The relative probability of vowels falling within the target F1 × F2 space was greater for HCs but did not differ with stress. Stress production resulted in greater listener vowel identification accuracy for speakers with mild dysarthria. Contrastive stress affected vowel formants for both groups. Perceptual results suggest that some speakers with dysarthria may benefit from a contrastive stress strategy to improve vowel intelligibility.

The Identification of High-pitched Sung Vowels in Sense and Nonsense Words by Professional Singers and Untrained Listeners

High-pitched sung vowels may be considered phonetically "underspecified" because of (i) the tuning of the F1 to the f0 accompanying pitch raising and (ii) the wide harmonic spacing of the voice source resulting in the undersampling of the vocal tract transfer function. Therefore, sung vowel intelligibility is expected to decrease as the f0 increases. Based on the literature of speech perception, it is often suggested that sung vowels are better perceived if uttered in consonantal (CVC) context than in isolation even at high f0. The results for singing, however, are contradictory. In the present study, we further investigate this question. We compare vowel identification in sense and nonsense CVC sequences and show that the positive effect of the context disappears if the number of legal choices in a perception test is similar in both conditions, meaning that any positive effect of the CVC context may only stem from the smaller number of possible responses, i.e., from higher probabilities. Additionally, it is also tested whether the training in production (i.e., singing training) may also lead to a perceptual advantage of the singers over nonsingers in the identification of high-pitched sung vowels. The results show no advantage of this kind.

Vowel Intelligibility in Children With and Without Dysarthria

Children with dysarthria due to cerebral palsy (CP) present with decreased vowel space area and reduced word intelligibility. Although a robust relationship exists between vowel space and word intelligibility, little is known about the intelligibility of vowels in this population. This exploratory study investigated the intelligibility of American English (AE) vowels produced by children with dysarthria and typically developing children (TD). Three CP and five TD repeated words with contrastive vowels /i-ɪ/, /æ-ɛ/, /ɑ-ʌ/, /o-u/ produced by a native AE adult. Adult listeners transcribed the utterances orthographically and rated their ease of understanding. Overall, CP presented with less-intelligible vowels than TD. For CP, a trend was found with the lowest intelligibility for /ɑ/ (CP = 7%, TD = 66%), /ɪ/ (CP = 30%, TD = 82%), and /ʌ/ (CP = 38%, TD = 99%), and more heterogeneous vowel confusions; however, intelligibility differences between vowels did not reach statistical significance. Clinical implications include that, unless further studies show vowel-specific effects, treatment targeting the entire vowel system may be warranted for increasing intelligibility.

Acoustic and perceptual correlates of subjectively rated sentence clarity in clear and conversational speech

Young adults with normal hearing and older adults with hearing loss performed subjective ratings of speech clarity on sentences spoken by all 41 talkers of the Ferguson Clear Speech Database. The sentences were selected from the CID Everyday Sentence lists and were produced under instructions to speak in a conversational manner and in a clear speaking style. A different set of 14 sentences was recorded in each style. Rated clarity will be compared between the two listener groups as well as among subgroups of talkers who differ in demographic and other characteristics. Clarity data will also be analyzed in conjunction with perceptual and acoustic data obtained in other investigations to reveal the relationship between vowel intelligibility and sentence clarity as well as the acoustic features that underlie perceived sentence clarity for different listener groups.

Sensitivity of odd-harmonic amplitudes to open quotient and skewing quotient in glottal airflow.

It is well known that a half-sinusoid has no odd harmonics other than the fundamental. If glottal flow in phonation were to approximate this exact waveshape, which is generally unlikely, some misperception of pitch and loss of vowel intelligibility would occur. The sensitivity of the glottal waveshape to this special shape is explored by systematically varying two parameters, open quotient and skewing quotient. Mild asymmetry (open quotient below 0.45 or above 0.55 and/or skewing quotient greater than 2.0) equalizes the odd-even harmonic series. Singers and speakers avoid the exact symmetry by skewing the flow pulse with source-filter interaction.

Using multiple measures to document change in English vowels produced by Japanese, Korean, and Spanish speakers: the case for goodness and intelligibility.

This study examined the effectiveness of using goodness ratings and intelligibility scores to document changes in vowel production following pronunciation training. The relationship between listener perceptions of goodness and intelligibility was also examined. Fifteen English language learner speakers (5 Japanese, 5 Korean, and 5 Spanish) participated in 16 sessions of vowel-focused pronunciation training. Pre- and posttraining judgments of 10 English vowels in /hVt/ context were conducted by 25 monolingual English speakers who served as listeners. Listeners judged vowel intelligibility using a 10-alternative forced-choice task and rated goodness using a 5-point Likert scale. Goodness ratings and intelligibility scores captured improvement in the accuracy of several vowels following training. However, some vowels that received better mean intelligibility scores received poorer mean goodness ratings following training. The relationship between goodness ratings and intelligibility scores revealed that vowels such as /æ/ and /ʌ/ were more dependent on goodness for intelligibility than vowels such as /i/ and /e/, which were highly intelligible even when they received poor goodness ratings. English vowels differ with respect to the importance of goodness for accurate identification by listeners. As such, clinicians should examine both goodness and intelligibility when measuring change following pronunciation training.

Vowel intelligibility and the second-language learner

Diane Kewley-Port’s work has contributed to our understanding of vowel perception and production in a wide variety of ways, from mapping the discriminability of vowel formants in conditions of minimal uncertainty to vowel processing in challenging conditions, such as increased presentation rate and noise. From the results of these studies, we have learned much about the limits of vowel perception for normal-hearing listeners and the robustness of vowels in speech perception. Continuously intertwined with this basic research has been its application to our understanding of vowel perception and vowel acoustics across various challenges, such as hearing impairment and second-language learning. Diane’s work on vowel perception and production by second-language learners and ongoing research stemming from her influence will be considered in light of several factors affecting communicative success and challenge for second-language learners. In particular, we will compare the influence of speaking style, noise, and syllable disruption on the intelligibility of vowels perceived and produced by native and non-native English-speaking listeners.

Intelligibility of American English Vowels and Consonants Spoken by International Students in the United States

PURPOSE The purpose of this study was to examine the intelligibility of English consonants and vowels produced by Chinese-native (CN), and Korean-native (KN) students enrolled in American universities. METHOD 16 English-native (EN), 32 CN, and 32 KN speakers participated in this study. The intelligibility of 16 American English consonants and 16 vowels spoken by native and nonnative speakers of English was evaluated by EN listeners. All nonnative speakers also completed a survey of their language backgrounds. RESULTS Although the intelligibility of consonants and diphthongs for nonnative speakers was comparable to that of native speakers, the intelligibility of monophthongs was significantly lower for CN and KN speakers than for EN speakers. Sociolinguistic factors such as the age of arrival in the United States and daily use of English, as well as a linguistic factor, difference in vowel space between native (L1) and nonnative (L2) language, partially contributed to vowel intelligibility for CN and KN groups. There was no significant correlation between the length of U.S. residency and phoneme intelligibility. CONCLUSION Results indicated that the major difficulty in phonemic production in English for Chinese and Korean speakers is with vowels rather than consonants. This might be useful for developing training methods to improve English intelligibility for foreign students in the United States.

Vocoder Simulations of Highly Focused Cochlear Stimulation With Limited Dynamic Range and Discriminable Steps

There is recent interest in focused stimulation of the cochlea via modalities such as tripolar electrical and infrared neural stimulation to improve speech in noise comprehension and music perception. The purpose of this work was to use vocoder-based simulations to investigate speech recognition for broad stimulation (standard monopolar paradigm) versus more focused stimulation under a variety of signal-to-noise ratios, dynamic ranges, and numbers of discriminable loudness steps. Vocoder simulations were used to assess the intelligibility of sentences, consonants, and vowels that were noise vocoded and presented to 7 normal-hearing listeners for identification. A novel aspect of the simulations presented here was the use of nonuniform quantization steps within the dynamic range to more closely simulate the Weber functions observed in cochlear implant users. Intelligibility was assessed for the different filter slopes under a variety of signal-to-noise ratio levels, dynamic ranges, and numbers of discriminable steps. Speech processed via vocoder simulations representing focused stimulation was found to be substantially more intelligible than speech processed via a monopolar electric vocoder simulation, with differences of up to 60 percentage points. There were no significant differences, however, seen between the two focused approaches (signal attenuations of 10 and 17 dB/mm) for the conditions investigated. Speech processed via the highly focused vocoder (17 dB/mm) was robust to constraints on small envelope dynamic range and small number of discriminable steps within the dynamic range, as high performance was maintained with at least a 5 dB dynamic range and eight or more discriminable steps. Significant drops in intelligibility were noted when the number of steps fell below eight. Highly focused stimulation-tripolar electrical and infrared neural stimulation-has potential for increased performance in noise compared with monopolar stimulation, but much work remains to bear this potential out and to take full advantage of each modality's strengths.

Intelligibility of American English vowels of native and non-native speakers in quiet and speech-shaped noise

This study examined intelligibility of twelve American English vowels produced by English, Chinese, and Korean native speakers in quiet and speech-shaped noise in which vowels were presented at six sensation levels from 0 dB to 10 dB. The slopes of vowel intelligibility functions and the processing time for listeners to identify vowels were dependent on speakers’ language backgrounds and non-native speakers’ vowel intelligibility in quiet. These results indicated that noise background affected non-native speakers’ vowel intelligibility more greatly than native speakers, possibly due to the acoustic deviations in non-native speech and lack of listeners’ experience to non-native produced speech.

The role of vowel inherent spectral change in the intelligibility of English vowels spoken by English-, Chinese-, and Korean-native speakers

This study is designed to investigate the relationship between Vowel Inherent Spectral Change (VISC) and vowel intelligibility of native and non-native speakers of English. VISC refers to the relatively slow varying changes in formant frequencies associated with each vowel category (Neary and Assman, 1986). Such spectral change has been known to be a major factor in the perception of both the phonetic and phonemic English vowels. In the previous research projects conducted in our lab, we recorded 12 English vowels in /hVd/ format spoken by English native (EN), Chinese native (CN) and Korean native (KN) speakers and examined vowel intelligibility. Overall, vowel intelligibility was significantly higher for native talkers than for non-native talkers. Within non-native speaker groups, CN had slightly higher intelligibility than KN speakers. In this study, we are going to analyze an acoustic feature of the vowels, VISC, spoken by these native and non-native speakers. It is hypothesized that the acoustic differences of vowels spoken by different groups of speakers can account for the variances in vowel intelligibility.