東華大學圖書館 |

Development of Singapore Mandarin Word Recognition Test Materials /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Development of Singapore Mandarin Word Recognition Test Materials // Soh Wanxian Kimberly.
作者:	Kimberly, Soh Wanxian,
面頁冊數:	1 electronic resource (201 pages)
附註:	Source: Dissertations Abstracts International, Volume: 86-06, Section: A.
Contained By:	Dissertations Abstracts International86-06A.
標題:	Reimbursement. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31777783
ISBN:	9798346780083

Development of Singapore Mandarin Word Recognition Test Materials /
Kimberly, Soh Wanxian,

Development of Singapore Mandarin Word Recognition Test Materials /Soh Wanxian Kimberly. - 1 electronic resource (201 pages)

Source: Dissertations Abstracts International, Volume: 86-06, Section: A.

Background: Speech recognition tests in the audiological test battery are useful for differential diagnoses and quantifying the benefit of hearing devices. They must be conducted in a listener's native language, so that obtained speech recognition scores (SRS) are not artificially lowered due to unfamiliarity with the language. Test materials should also be phonemically balanced, contain lists that are perceptually equivalent, and have good testretest reliability. Existing English materials validated on monolingual native English speakers cannot be used to test on Mandarin monolinguals or Mandarin-English bilinguals in Singapore. Mandarin speech recognition test materials developed in China and Taiwan are unsuitable to be adapted for local use, as China Mandarin and Taiwan Mandarin have significant phonological, phonetic, and lexical differences from Singapore Mandarin. Although speech recognition test materials have also been developed in Singapore Mandarin (SC-10), the materials were not phonemically balanced or perceptually equivalent.Objective: The purpose of this study was to develop and validate a set of phonemically balanced Singapore Mandarin speech recognition test materials.Methods: Test materials were developed over five studies. All participants, speakers and evaluators were native speakers of Singapore Mandarin. 1) Commonly heard Singapore Mandarin words were extracted from subtitles of local variety programmes. The 500 most frequently occurring characters were rated for familiarity by 50 participants. 2) Familiar characters from Study 1 were used in the creation of monosyllabic, phonemically balanced 25-item word lists. 3) The characters were recorded into syllables by a male and a female speaker. Four evaluators examined the recordings for intelligibility and naturalness. 4) Psychometric curves were established for each recording from each speaker with 20 normal-hearing participants. For each syllable, the recording of the speaker with the steeper slope on the psychometric curve was selected and arranged into word lists. List equivalence was evaluated via repeated measures ANOVA and lists that differed significantly were eliminated. 5) Remaining word lists were validated on 38 normal-hearing and 25 hearing-impaired participants.Results: 1) 467 characters were rated as familiar, which equated to 389 unique syllables due to the presence of homophones. 2) Of the 389 syllables, 256 were used in the creation of 16 word lists. 3) 253 male speaker's tokens and 247 female speaker's tokens met the intelligibility criteria. 4) 8 of the 16 word lists were eliminated as they were not perceptually equivalent to the other lists. 5) Reference speech recognition curves were established. The slope of the curve at 50% was 4.09%/dB for normal-hearing participants and 3.01%/dB for hearing-impaired participants. Test and retest SRS had a mean difference of 2.23%. Multifactorial ANOVA results indicated no list effect on SRS. Conclusion: This study used a novel method of creating speech recognition test materials from a collection of Mandarin programme subtitles, and using recordings from two different speakers. Results from the final validation phase show values which are comparable to similar existing materials. The study has addressed the limitations of existing test materials and has found the developed Singapore Mandarin speech recognition test materials to be suitable for clinical use.

English

ISBN: 9798346780083Subjects--Topical Terms:

3680791
Reimbursement.

Development of Singapore Mandarin Word Recognition Test Materials /
LDR:04960nmm a22004693i 4500 001 2435334
005 20251001074916.5
006 m o d
007 cr|nu||||||||
008 261011s2023 miu||||||m |||||||eng d
020 $a 9798346780083
035 $a (MiAaPQD)AAI31777783
035 $a AAI31777783
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Kimberly, Soh Wanxian, $e author. $3 3825621
245 1 0 $a Development of Singapore Mandarin Word Recognition Test Materials / $c Soh Wanxian Kimberly.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 1 electronic resource (201 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-06, Section: A.
500 $a Advisors: Jenny, Loo Hooi Yin; Martin, William Hal; Lena, Wong Lai Nar.
502 $b Ph.D. $c National University of Singapore (Singapore) $d 2023.
520 $a Background: Speech recognition tests in the audiological test battery are useful for differential diagnoses and quantifying the benefit of hearing devices. They must be conducted in a listener's native language, so that obtained speech recognition scores (SRS) are not artificially lowered due to unfamiliarity with the language. Test materials should also be phonemically balanced, contain lists that are perceptually equivalent, and have good testretest reliability. Existing English materials validated on monolingual native English speakers cannot be used to test on Mandarin monolinguals or Mandarin-English bilinguals in Singapore. Mandarin speech recognition test materials developed in China and Taiwan are unsuitable to be adapted for local use, as China Mandarin and Taiwan Mandarin have significant phonological, phonetic, and lexical differences from Singapore Mandarin. Although speech recognition test materials have also been developed in Singapore Mandarin (SC-10), the materials were not phonemically balanced or perceptually equivalent.Objective: The purpose of this study was to develop and validate a set of phonemically balanced Singapore Mandarin speech recognition test materials.Methods: Test materials were developed over five studies. All participants, speakers and evaluators were native speakers of Singapore Mandarin. 1) Commonly heard Singapore Mandarin words were extracted from subtitles of local variety programmes. The 500 most frequently occurring characters were rated for familiarity by 50 participants. 2) Familiar characters from Study 1 were used in the creation of monosyllabic, phonemically balanced 25-item word lists. 3) The characters were recorded into syllables by a male and a female speaker. Four evaluators examined the recordings for intelligibility and naturalness. 4) Psychometric curves were established for each recording from each speaker with 20 normal-hearing participants. For each syllable, the recording of the speaker with the steeper slope on the psychometric curve was selected and arranged into word lists. List equivalence was evaluated via repeated measures ANOVA and lists that differed significantly were eliminated. 5) Remaining word lists were validated on 38 normal-hearing and 25 hearing-impaired participants.Results: 1) 467 characters were rated as familiar, which equated to 389 unique syllables due to the presence of homophones. 2) Of the 389 syllables, 256 were used in the creation of 16 word lists. 3) 253 male speaker's tokens and 247 female speaker's tokens met the intelligibility criteria. 4) 8 of the 16 word lists were eliminated as they were not perceptually equivalent to the other lists. 5) Reference speech recognition curves were established. The slope of the curve at 50% was 4.09%/dB for normal-hearing participants and 3.01%/dB for hearing-impaired participants. Test and retest SRS had a mean difference of 2.23%. Multifactorial ANOVA results indicated no list effect on SRS. Conclusion: This study used a novel method of creating speech recognition test materials from a collection of Mandarin programme subtitles, and using recordings from two different speakers. Results from the final validation phase show values which are comparable to similar existing materials. The study has addressed the limitations of existing test materials and has found the developed Singapore Mandarin speech recognition test materials to be suitable for clinical use.
546 $a English
590 $a School code: 1883
650 4 $a Reimbursement. $3 3680791
650 4 $a Quantitative psychology. $3 2144748
650 4 $a Voice recognition. $3 3564741
650 4 $a Audiometry. $3 596229
650 4 $a Hearing loss. $3 3697149
650 4 $a Ostomy. $3 3825622
650 4 $a International organizations. $3 1998637
650 4 $a Subtitles & subtitling. $3 3688954
650 4 $a Deafness. $3 583177
650 4 $a Audiology. $3 537237
650 4 $a Speaking. $3 3561999
650 4 $a Speech. $3 530580
650 4 $a Chinese languages. $3 3560871
650 4 $a Disability studies. $3 543687
650 4 $a Language. $3 643551
650 4 $a Southeast Asian studies. $3 3344898
650 4 $a Surgery. $3 707153
650 4 $a Translation studies. $3 3310485
690 $a 0632
690 $a 0300
690 $a 0201
690 $a 0679
690 $a 0454
690 $a 0703
690 $a 0222
690 $a 0576
690 $a 0213
710 2 $a National University of Singapore (Singapore). $e degree granting institution. $3 3765631
720 1 $a Jenny, Loo Hooi Yin $e degree supervisor.
720 1 $a Martin, William Hal $e degree supervisor.
720 1 $a Lena, Wong Lai Nar $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-06A.
790 $a 1883
791 $a Ph.D.
792 $a 2023
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31777783