Psychometrically equivalent trisyllabic words for speech reception threshold testing in Mandarin

Chinese Emotional Speech Audiometry Project (CESAP): Establishment and Validation of a New Material Set With Emotionally Neutral Disyllabic Words

PURPOSE

The Chinese Emotional Speech Audiometry Project (CESAP) aims to establish a new material set for Chinese speech audiometry tests, which can be used in both neutral and emotional prosody settings. As the first endeavor of CESAP, this study demonstrates the development of the material foundation and reports its validation in neutral prosody.

METHOD

In the development step, 40 phonetically balanced word lists consisting of 30 Chinese disyllabic words with neutral valence were first generated. In a following affective rating experiment, 35 word lists were qualified for validation based on the familiarity and valence ratings from 30 normal-hearing (NH) participants. For validation, performance-intensity functions of each word list were fitted with responses from 60 NH subjects under six presentation levels (-1, 3, 5, 7, 11, and 20 dB HL). The final material set was determined by the intelligibility scores at each decibel level and the mean slopes.

RESULTS

First, 35 lists satisfied the criteria of phonetic balance, limited repetitions, high familiarity, and neutral valence and were selected for validation. Second, 15 lists were compiled in the final material set based on the pairwise differences in intelligibility scores and the fitted 20%-80% slopes. The established material set had high reliability and validity and was sensitive to detect intelligibility changes (50% slope: 6.20%/dB; 20%-80% slope: 5.45%/dB), with small covariance of variation for thresholds (15%), 50% slope (12%), and 20%-80% slope (12%).

CONCLUSION

Our final material set of 15 word lists takes the initiative to control the emotional aspect of audiometry tests, which enriches available Mandarin speech recognition materials and warrants future assessments in emotional prosody among populations with hearing impairments.

SUPPLEMENTAL MATERIAL

https://doi.org/10.23641/asha.25742814.

Psychometrically Equivalent Fante Speech Recognition Threshold Materials by a Female Talker

PURPOSE

This study aimed to create and evaluate digitally recorded speech materials for speech recognition threshold (SRT) testing among children and adults in Fante. Fifty-one (51) familiar trisyllabic words were chosen from a list of 107 widely used trisyllabic Fante words. They were digitally captured and edited to create the same root mean square as a 1-kHz calibration tone.

METHOD

The study used a three-phase cross-sectional study method. Twenty native Fante speakers with normal hearing thresholds were selected at random for listener evaluation. For each of the words, the researchers used logistic regression to measure the slope, intercepts, and psychometric function slope at 50% and from 20% to 80%. In the study, the intensity of each word was modified digitally, so that the threshold at 50% of each word was equal to the mean pure-tone average (PTA) of the participants to increase the homogeneity of the thresholds of the selected words.

RESULTS

A final list of 25 familiar homogenous words with the same tone patterns of slopes greater than 7%/dB was finally selected and recorded for speech audiometry in Fante.

CONCLUSIONS

Psychometrically equivalent trisyllabic words in Fante were successfully developed and evaluated for SRT testing in Ghana. There is a need for the development of speech audiometry materials in other Ghanaian languages.

Validating Four Hmong Word Recognition Tests With Normal-Hearing Bilingual Hmong Individuals

Purpose:

The purpose of this study was to validate four digitally recorded and phonetically balanced 50-word recognition lists in the White Hmong dialect with normal-hearing bilingual Hmong adults.

Method:

Using a randomized, incomplete-block design, each participant listened to and repeated four unique Hmong lists delivered by a female and a male talker. Participants were also tested with an English word list—List 1A of the Northwestern University Auditory Test No. 6. Participants' correct pronunciation of each word was scored. A nonparametric Mann–Whitney U Location Difference Test for Equivalence using two one-sided tests equivalence hypothesis: −0.02 < [(List_1) – (List_2)] < 0.02 was conducted to assess equivalence among all four Hmong and the English lists.

Results:

Seventy Hmong speakers participated in this study (35 women, 35 men; M age = 29.5 years, SD = 7.1). In all four Hmong lists, 93.5% (187/200) words met the validation criteria for ≥ 92% correct pronunciation. The 13 difficult words were deemed adequate by a Hmong panel and, therefore, were included to maintain four unique, balanced word lists. The test revealed that the Hmong and English word lists were considered equivalent at the 2% bound.

Conclusion:

The four Hmong word lists were validated to ensure an equal range of word difficulty across the lists.

Development Of Speech Material For An Armenian Speech Recognition Threshold Test

Introduction — For the Armenian language, no standardized audiometric speech perception test is available. The purpose of this research was to develop, digitally record and evaluate an Armenian multisyllabic speech audiometry test, which can be used to measure speech recognition thresholds in Armenian native speakers. Material and Methods — To create a homogeneous multisyllabic speech corpus, Armenian numerals from 10-100 with 2-4 syllables were selected as general sample and digitally recorded by a female native Armenian speaker. For equalizing the speech recognition threshold between the test items, the speech discrimination function for each numeral was subsequently evaluated by five normal hearing native Armenian listeners in an experimental study. Results — Based on the phonemic structure of the Armenian language, 20 phonemically homogeneous test lists were created. The phoneme distributions of each test list correlated significantly and positively with that of the general sample (all Pearson moment correlation coefficients >0.960; all ps <0.001). Comparison of the phoneme distributions of test lists to that of the Armenian language showed that the test lists represent the language corpus quite well. After adjusting for actual threshold levels, speech discrimination functions are comparable between all numerals used. Conclusion — The developed test lists are a phonetically homogenous representation of the Armenian language and serve as an appropriated base for future clinical measurements of speech recognition threshold in Armenian speaking listeners.

Development of the Mandarin Nonsense Word Identification Test

OBJECTIVE

This study aimed to develop a digitised Mandarin Nonsense Word Speech Perception Test for use in Malaysia, a multilingual country in Southeast Asia.

DESIGN

In Phase I, 400 vowel-consonant-vowel (VCV) nonsense word samples containing 20 Mandarin consonants in /a/, /i/, or /u/ contexts were recorded from two speakers of different genders. Acoustic analyses, sound quality ratings, and item validations were used to guide selection of items to form two gender-specific test lists. In Phase II, performance-intensity functions and test-retest reliability for the lists were established.

STUDY SAMPLE

Native Mandarin-speaking adults with normal hearing participated in Phase I (n = 10) and Phase II (n = 69).

RESULTS

Eighty-four of the 400 VCV words were selected to form two gender-specific test lists. A two-way repeated measure ANOVA revealed a significant interaction effect between speaker-gender and presentation level [F (4.88, 283.20) = 22.79, p < 0.001, η_p²= 0.28]. Intraclass correlation scores of 0.75 and 0.87 were obtained for the female-speaker and male-speaker lists respectively.

CONCLUSIONS

The preliminary normative data of the Mandarin nonsense word test have been developed. It is recommended to use separate gender-specific norms when conducting the test. The test has good validity and reliability for testing Mandarin-speaking adults in Malaysia.

The validity of an isiZulu speech reception threshold test for use with adult isiZulu speakers

Background

This study continued the development of an isiZulu speech reception threshold (zSRT) test for use with first language, adult speakers of isiZulu.

Objectives

The objective of this study was to determine the convergent and concurrent validity of the zSRT test.

Methods

One hundred adult isiZulu first-language speakers with normal hearing and 76 first-language, adult isiZulu speakers with conductive or sensorineural hearing losses ranging from mild to severe were assessed on pure tone audiometry and a newly developed isiZulu SRT test. Convergent validity was established through agreement of the zSRT scores with pure tone average (PTA) scores. Concurrent validity was assessed by examining the steepness of the psychometric curve for each word in the zSRT test for each type and degree of hearing loss.

Results

Intraclass correlation coefficient analyses showed zSRT scores were in substantial to very high agreement with PTA scores for the normal hearing and hearing loss groups (NH – right ear ICC consistency = 0.78, left ear ICC = 0.67; HL – right ear ICC consistency = 0.97, left ear ICC consistency = 0.95). The mean psychometric slope (%/dB) at 50% correct perception for all words in the zSRT test was 4.92%/dB for the mild conductive hearing loss group, 5.26%/dB for the moderate conductive hearing loss group, 2.85%/dB for the moderately severe sensorineural hearing loss group and 2.47%/dB for the severe sensorineural hearing loss group. These slopes were appropriate for the degree of hearing loss observed in each group.

Conclusion

The zSRT test showed convergent and concurrent validity for assessing SRT in first language, adult speakers of isiZulu.

The characteristics of monosyllable recognition in Mandarin-speaking patients with auditory neuropathy

Background: A set of Chinese Mandarin monosyllable test lists has been widely used in clinical diagnosis, while the performance of the Mandarin-speaking patients with auditory neuropathy (AN) in the monosyllable tests was still unknown.Objectives: To analyze the characteristics of monosyllable recognition in Mandarin-speaking patients with AN.Materials and methods: Nineteen Mandarin-speaking patients diagnosed with AN were recruited to obtain the performance-intensity (P-I) functions. They were divided into two paired groups according to two conditions: (1) the rising-slope (RS) and the non-rising-slope pattern (NRS) group for the audiogram; or (2) the male and the female group for gender.Results: The performance at 20 and 30 dB SL showed negative correlation with the pure tone-averaged thresholds of 0.5, 1.0, 2.0 and 4.0 kHz (4FA). We found significant differences between the P-I functions of the RS and the NRS group, and surprisingly found that between the male and female as well although there was no difference in 4FA between the paired groups. A notable rollover phenomenon in P-I functions was detected in each group.Conclusions: The monosyllable recognition scores of Mandarin-speaking AN patients showed numerous particular characteristics reflected by P-I functions. Acquiring the complete P-I functions at multiple presentation levels and the rollover index may give us more information about the diagnosis and prognosis.

The internal and external consistency of a speech reception threshold test for isiZulu speakers with normal hearing sensitivity

Background and objectives

This study investigated reliability, particularly the internal and external consistency, of a new isiZulu speech reception threshold (SRT) test.

Methods

To examine internal consistency, 21 adult isiZulu speakers with normal hearing sensitivity completed the SRT test using the first and second halves of the SRT wordlist in the same test session. To examine external consistency, a separate 23 adult isiZulu speakers with normal hearing sensitivity completed the SRT test, using the whole word list on two occasions 4 weeks apart. Consistency of SRT test scores in these test conditions was measured using intraclass correlation coefficient analyses (a measure of the consistency or reproducibility of different observations of the same quantity) and Bland and Altman analyses of agreement (a comparison of measurement error with the expected variation amongst subjects).

Results

Intraclass correlation coefficient values ranged from 0.69 to 0.79, showing the isiZulu test scores were highly consistent between the test and retest conditions used in this study. Bland and Altman analyses showed that isiZulu speakers with normal hearing sensitivity can be expected to return isiZulu SRT test scores that differ by no more than 7.5 dB HL – 8.7 dB HL between original and repeat assessments.

Conclusion

The isiZulu SRT test was reliable, showing high internal and external consistency, when used to assess first-language speakers of isiZulu with normal hearing sensitivity. These findings warrant continued development of the isiZulu SRT test for eventual clinical use. This development should include validating this test on first-language speakers of isiZulu with and without hearing loss.

First-language raters' opinions when validating word recordings for a newly developed speech reception threshold test

Background

The purpose of this study was to consider the value of adding first-language speaker ratings to the process of validating word recordings for use in a new speech reception threshold (SRT) test in audiology. Previous studies had identified 28 word recordings as being suitable for use in a new SRT test. These word recordings had been shown to satisfy the linguistic criteria of familiarity, phonetic dissimilarity and tone, and the psychometric criterion of homogeneity of audibility.

Objectives

The aim of the study was to consider the value of adding first-language speakers’ ratings when validating word recordings for a new SRT test.

Method

A single observation, cross-sectional design was used to collect and analyse quantitative data in this study. Eleven first-language isiZulu speakers, purposively selected, were asked to rate each of the word recordings for pitch, clarity, naturalness, speech rate and quality on a 5-point Likert scale. The percent agreement and Friedman test were used for analysis.

Results

More than 20% of these 11 participants rated the three-word recordings below ‘strongly agree’ in the category of pitch or tone, and one-word recording below ‘strongly agree’ in the categories of pitch or tone, clarity or articulation and naturalness or dialect.

Conclusion

The first-language speaker ratings proved to be a valuable addition to the process of selecting word recordings for use in a new SRT test. In particular, these ratings identified potentially problematic word recordings in the new SRT test that had been missed by the previously and more commonly used linguistic and psychometric selection criteria.

Comparison of the South African Spondaic and CID W-1 wordlists for measuring speech recognition threshold

BACKGROUND

The home language of most audiologists in South Africa is either English or Afrikaans, whereas most South Africans speak an African language as their home language. The use of an English wordlist, the South African Spondaic (SAS) wordlist, which is familiar to the English Second Language (ESL) population, was developed by the author for testing the speech recognition threshold (SRT) of ESL speakers.

OBJECTIVES

The aim of this study was to compare the pure-tone average (PTA)/SRT correlation results of ESL participants when using the SAS wordlist (list A) and the CID W-1 spondaic wordlist (list B - less familiar; list C - more familiar CID W-1 words).

METHOD

A mixed-group correlational, quantitative design was adopted. PTA and SRT measurements were compared for lists A, B and C for 101 (197 ears) ESL participants with normal hearing or a minimal hearing loss (<26 dBHL; mean age 33.3).

RESULTS

The Pearson correlation analysis revealed a strong PTA/SRT correlation when using list A (right 0.65; left 0.58) and list C (right 0.63; left 0.56). The use of list B revealed weak correlations (right 0.30; left 0.32). Paired sample t-tests indicated a statistically significantly stronger PTA/SRT correlation when list A was used, rather than list B or list C, at a 95% level of confidence.

CONCLUSIONS

The use of the SAS wordlist yielded a stronger PTA/SRT correlation than the use of the CID W-1 wordlist, when performing SRT testing on South African ESL speakers with normal hearing, or minimal hearing loss (<26 dBHL).

The Effect of Dialect on Speech Audiometry Testing

Purpose

In this study, the authors examined the validity of using materials from 2 nonregional yet mutually intelligible dialects to evaluate an individual's speech recognition threshold (SRT) and word recognition (WR) abilities and whether a speaker of 1 dialect could accurately administer and score materials in the other dialect.

Method

Previously created SRT and WR materials were presented to 32 Mandarin listeners with normal hearing: 16 speakers of Mainland Mandarin and 16 speakers of Taiwan Mandarin. Hearing abilities were examined using SRT and WR materials created for speakers from 2 different regional dialects. Presentation of the materials occurred during 2 test sessions, counterbalanced across material and listener dialect. Listener responses were evaluated by 2 judges; 1 spoke Mainland Mandarin, and the other spoke Taiwan Mandarin.

Results

For the SRT and WR results, differences in listener performance were statistically significant across material and listener dialect, with threshold differences of less than 2 dB HL when collapsed across session. The interscorer percentage of agreement was 99.5% for SRT and 99.1% for WR testing.

Conclusion

Testing with materials in a different regional dialect does have a measurable impact on SRT and WR performance. However, this difference, though reliable, is small enough to have a negligible impact on clinical findings.

Development and validation of the Mandarin disyllable recognition test

CONCLUSION

Given the phonetic balancing across lists and the validation with spectrally degraded speech, the present Mandarin disyllable recognition test (DRT) materials may be useful for assessing speech performance of Mandarin-speaking cochlear implant (CI) users. If combined with the previously developed sentence materials, these materials would help to establish standardized speech perception tests for Mandarin-speaking hearing-impaired (HI) and CI patients.

OBJECTIVES

To develop standardized Mandarin DRT materials that can be used to evaluate the speech performance of Mandarin-speaking HI and CI patients, and to establish standardized Mandarin speech perception test materials that include both disyllables and sentences.

METHODS

Ten phonetically balanced Mandarin DRT lists were developed. The DRT materials were validated in 8 normal-hearing (NH) subjects listening to unprocessed speech and in 10 NH subjects listening to a 4-channel, sine-wave vocoded acoustic simulation of CI speech processing. Performance with the DRT materials was compared to that with Mandarin sentence materials previously developed by our group.

RESULTS

The distribution of vowels, consonants, and tones within each DRT list was similar to that observed across commonly used Chinese characters. There was no significant difference in disyllable word recognition across lists in both unprocessed and four-channel vocoded speech. There was a significant correlation between disyllable and sentence recognition performance.

The role of selective attention in the acquisition of English tense and lax vowels by native Spanish listeners: comparison of three training methods

This study investigates the role of two processes, cue enhancement (learning to attend to acoustic cues which characterize a speech contrast for native listeners) and cue inhibition (learning to ignore cues that do not), in the acquisition of the American English tense and lax ([i] vs.[I]) vowels by native Spanish listeners. This contrast is acoustically distinguished by both vowel spectrum and duration. However, while native English listeners rely primarily on spectrum, inexperienced Spanish listeners tend to rely exclusively on duration. Twenty-nine native Spanish listeners, initially reliant on vowel duration, received either enhancement training, inhibition training, or training with a natural cue distribution. Results demonstrated that reliance on spectrum properties increased over baseline for all three groups. However, inhibitory training was more effective relative to enhancement training and both inhibitory and enhancement training were more effective relative to natural distribution training in decreasing listeners' attention to duration. These results suggest that phonetic learning may involve two distinct cognitive processes, cue enhancement and cue inhibition, that function to shift selective attention between separable acoustic dimensions. Moreover, cue-specific training (whether enhancing or inhibitory) appears to be more effective for the acquisition of second language speech contrasts.

Development of speech reception threshold materials for speakers of Taiwan Mandarin

The aim of this research was to develop, digitally record, evaluate, and equate speech audiometry materials that can be used to measure the speech reception threshold (SRT) in quiet for native speakers of Taiwan Mandarin. Familiar trisyllabic words were digitally recorded by male and female talkers of Taiwan Mandarin and subsequently evaluated by 20 native listeners with normal hearing at 14 intensity levels (-10 to 16 dB HL) in 2 dB increments. Using logistic regression, psychometric functions were calculated for all words. Twenty-eight words with comparatively steep psychometric functions were selected and digitally adjusted to match the mean subject pure-tone average (5.0 dB HL). This resulted in a list of words that are relatively homogeneous in threshold audibility and psychometric function slope. The mean slopes for the 28 selected male and female trisyllabic Taiwan Mandarin words were 11.3%/dB and 11.7%/dB, respectively.

The relationship between native allophonic experience with vowel duration and perception of the English tense/lax vowel contrast by Spanish and Russian listeners

Two studies explored the role of native language use of an acoustic cue, vowel duration, in both native and non-native contexts in order to test the hypothesis that non-native listeners' reliance on vowel duration instead of vowel quality to distinguish the English tense/lax vowel contrast could be explained by the role of duration as a cue in native phonological contrasts. In the first experiment, native Russian, Spanish, and American English listeners identified stimuli from a beat/bit continuum varying in nine perceptually equal spectral and duration steps. English listeners relied predominantly on spectrum, but showed some reliance on duration. Russian and Spanish speakers relied entirely on duration. In the second experiment, three tests examined listeners' use of vowel duration in native contrasts. Duration was equally important for the perception of lexical stress for all three groups. However, English listeners relied more on duration as a cue to postvocalic consonant voicing than did native Spanish or Russian listeners, and Spanish listeners relied on duration more than did Russian listeners. Results suggest that, although allophonic experience may contribute to cross-language perceptual patterns, other factors such as the application of statistical learning mechanisms and the influence of language-independent psychoacoustic proclivities cannot be ruled out.

Word Recognition Materials for Native Speakers of Taiwan Mandarin

Purpose

To select, digitally record, evaluate, and psychometrically equate word recognition materials that can be used to measure the speech perception abilities of native speakers of Taiwan Mandarin in quiet.

Method

Frequently used bisyllabic words produced by male and female talkers of Taiwan Mandarin were digitally recorded and subsequently evaluated using 20 native listeners with normal hearing at 10 intensity levels (−5 to 40 dB HL) in increments of 5 dB.

Results

Using logistic regression, 200 words with the steepest psychometric slopes were divided into 4 lists and 8 half-lists that were relatively equivalent in psychometric function slope. To increase auditory homogeneity of the lists, the intensity of words in each list was digitally adjusted so that the threshold of each list was equal to the midpoint between the mean thresholds of the male and female half-lists.

Conclusions

Digital recordings of the word recognition lists and the associated clinical instructions are available on CD upon request.

Psychometrically equivalent Russian speech audiometry materials by male and female talkers

Despite the large number of individuals who speak Russian, only a limited number of high-quality speech audiometry materials are available in a standard dialect of Russian. Thus, the purpose of this study was to develop and evaluate speech audiometry materials that can be used to measure word recognition and SRT testing in quiet for native speakers of Russian. Familiar monosyllabic and bisyllabic words were digitally recorded by male and female talkers of Russian and subsequently evaluated by native listeners. Using logistic regression, psychometric functions were then calculated for all words. Selected monosyllabic words were digitally adjusted to create word recognition lists which are relatively homogeneous with respect to audibility and psychometric slope. Speech reception threshold materials were developed by selecting twenty-five bisyllabic words with relatively steep psychometric function slopes (12.1%/dB and 9.9 %/dB) and digitally equating their intensity to match the mean PTA of the native listeners. Digital recordings of the resulting psychometrically equivalent speech audiometry materials are available on compact disc.