Journal of voice : official journal of the Voice Foundation · 2025
Exploring Objective Correlates to Perceived Severity in the Covarying Breathy, Rough, and Strained Voice Qualities of Dysphonic Voices.
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.
Awan Shaheen N, Park Yeonggwang, Anand Supraja, Shrivastav Rahul, Eddins David A
Аннотация
OBJECTIVE: While many research studies evaluate a single voice quality dimension to improve measurement precision, this approach often fails to capture the covariance among more than one voice quality. The goal of the current study was to investigate the relative utility of objective acoustic and bioinspired psychoacoustic-based measurement methods in a set of dysphonic voices that covary in severity from mild to severe along three voice quality dimensions.
METHODS: Stimuli included 26 voices that varied across three levels of dysphonic severity (mild, moderate, and severe) in three covarying voice quality dimensions (breathy, rough, and strain) based on the dimension-specific severity rankings of all 26 stimuli by four voice experts. Subsequent analyses of the vocal acoustic signals included two computations of the cepstral peak prominence (CPP) and three psychoacoustic-based methods derived from the front end of models of auditory processing, including pitch strength (PS), temporal envelope standard deviation, and spectral sharpness.
RESULTS: Analyses revealed that breathiness severity was strongly correlated with PS and CPPPRAAT, moderately correlated with CPPADSV, and not significantly correlated with the temporal envelope standard deviation measure. Roughness severity was significantly and moderately correlated with the temporal envelope standard deviation measure and CPPADSV. Both PS and sharpness were significantly correlated with strain severity. PS was the single most strongly correlated metric with overall severity.
CONCLUSIONS: The relative strength of acoustic and bioinspired psychoacoustic-based measures of the voice signal can substantially vary depending upon the perceptual focus of the raters that are rating/ranking a set of voice samples. While PS emerged as the single strongest correlate of ranked breathiness and strain, as well as ranked severity across breathy, rough, and strained voice types, other measures were observed to provide important contributions in marking the presence and severity of roughness (CPPADSV and temporal envelope standard deviation) and strain (sharpness).
METHODS: Stimuli included 26 voices that varied across three levels of dysphonic severity (mild, moderate, and severe) in three covarying voice quality dimensions (breathy, rough, and strain) based on the dimension-specific severity rankings of all 26 stimuli by four voice experts. Subsequent analyses of the vocal acoustic signals included two computations of the cepstral peak prominence (CPP) and three psychoacoustic-based methods derived from the front end of models of auditory processing, including pitch strength (PS), temporal envelope standard deviation, and spectral sharpness.
RESULTS: Analyses revealed that breathiness severity was strongly correlated with PS and CPPPRAAT, moderately correlated with CPPADSV, and not significantly correlated with the temporal envelope standard deviation measure. Roughness severity was significantly and moderately correlated with the temporal envelope standard deviation measure and CPPADSV. Both PS and sharpness were significantly correlated with strain severity. PS was the single most strongly correlated metric with overall severity.
CONCLUSIONS: The relative strength of acoustic and bioinspired psychoacoustic-based measures of the voice signal can substantially vary depending upon the perceptual focus of the raters that are rating/ranking a set of voice samples. While PS emerged as the single strongest correlate of ranked breathiness and strain, as well as ranked severity across breathy, rough, and strained voice types, other measures were observed to provide important contributions in marking the presence and severity of roughness (CPPADSV and temporal envelope standard deviation) and strain (sharpness).