Journal of voice : official journal of the Voice Foundation · 2019
Acoustic Voice Quality Index and Acoustic Breathiness Index: Analysis With Different Speech Material in the Brazilian Portuguese.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Englert Marina, Lima Livia, Behlau Mara
Аннотация
OBJECTIVE: To analyze the Acoustic Voice Quality Index (AVQI) and the Acoustic Breathiness Index (ABI) concurrent validity and diagnostic accuracy with different speech materials.
METHODS: Voices of 53 subjects (40 dysphonic; 13 vocally health) were recorded: vowel /a/ + counting numbers 1-20 (42 syllables) + reading text (138 syllables). Numbers and text were edited in order to achieve 3 seconds of voiced segments, such as the vowel /a/ (average of 18.81 and 32.49 syllables; confidence interval of 1.87 and 2.30). The audio files were edited to have 17 syllables for numbers and 32 for text. Three voice specialists perceptually judge the overall voice quality (G) and the breathiness (B). AVQI's and ABI's precision and concurrent validity were assessed.
RESULTS: The intra- and inter-rater reliability were high. Reading text presented higher concurrent validity (r) than automatic speech and excellent area under the receiver-operating characteristic curve for AVQI (0.963) and ABI (0.929). Counting numbers presented good area under the receiver-operating characteristic curve for AVQI (0.870) and excellent for ABI (0.924). Counting numbers produced higher sensitivity for ABI (95.2%) and reading text higher specificity for both indexes (AVQI = 100%; ABI = 90.90%). Reading text presented higher AVQI and ABI scores than numbers, therefore, reading seems to reveal more vocal deviations; however, perceptual judgment can be similar in both samples.
CONCLUSIONS: Different speech materials may impact acoustic outcomes and certain voice characteristics may not be evident. Reading text offers higher diagnostic accuracy. Clinician and/or researchers must select and standardize the speech sample according to their goals.
METHODS: Voices of 53 subjects (40 dysphonic; 13 vocally health) were recorded: vowel /a/ + counting numbers 1-20 (42 syllables) + reading text (138 syllables). Numbers and text were edited in order to achieve 3 seconds of voiced segments, such as the vowel /a/ (average of 18.81 and 32.49 syllables; confidence interval of 1.87 and 2.30). The audio files were edited to have 17 syllables for numbers and 32 for text. Three voice specialists perceptually judge the overall voice quality (G) and the breathiness (B). AVQI's and ABI's precision and concurrent validity were assessed.
RESULTS: The intra- and inter-rater reliability were high. Reading text presented higher concurrent validity (r) than automatic speech and excellent area under the receiver-operating characteristic curve for AVQI (0.963) and ABI (0.929). Counting numbers presented good area under the receiver-operating characteristic curve for AVQI (0.870) and excellent for ABI (0.924). Counting numbers produced higher sensitivity for ABI (95.2%) and reading text higher specificity for both indexes (AVQI = 100%; ABI = 90.90%). Reading text presented higher AVQI and ABI scores than numbers, therefore, reading seems to reveal more vocal deviations; however, perceptual judgment can be similar in both samples.
CONCLUSIONS: Different speech materials may impact acoustic outcomes and certain voice characteristics may not be evident. Reading text offers higher diagnostic accuracy. Clinician and/or researchers must select and standardize the speech sample according to their goals.