The Effect of Gender and Speech Task on Cepstral- and Spectral-Measures of Korean Normal Speakers

최성희,; 최철희,

doi:10.21848/asr.2016.12.3.157

Cited by 19 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several studies have investigated the impact of the speech task on acoustic parameters [24], [64], [65]. Our study tested the reproducibility of a wide range of voice parameters during counting and reading-a-story tasks.…”

Section: Discussionmentioning

confidence: 99%

“…Kiss and Vicsi reported that the measurement of speech features, mainly those calculated over sustained vowels or voiced parts of reading-a-story, is affected by the type of speech task [16]. It was also found that quantifying spectral and cepstral acoustic features, whether from vowel or continuous speech, is dependent on speech content [24]. A study of healthy speakers revealed that different speech types, such as counting, reading passages, and spontaneous speech, impacted the vibration frequency of the vocal folds in connected speech (speaking fundamental frequency) [25].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Reproducibility of Bio-Acoustic Features is Associated With Sample Duration, Speech Task, and Gender

Almaghrabi

Thewlis

Thwaites

et al. 2022

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Bio-acoustic properties of speech show evolving value in analyzing psychiatric illnesses. Obtaining a sufficient speech sample length to quantify these properties is essential, but the impact of sample duration on the stability of bio-acoustic features has not been systematically explored. We aimed to evaluate bio-acoustic features' reproducibility against changes in speech durations and tasks. We extracted source, spectral, formant, and prosodic features in 185 English-speaking adults (98 w, 87 m) for reading-a-story and counting tasks. We compared features at 25% of the total sample duration of the reading task to those obtained from non-overlapping randomly selected sub-samples shortened to 75%, 50%, and 25% of total duration using intraclass correlation coefficients. We also compared the features extracted from entire recordings to those measured at 25% of the duration and features obtained from 50% of the duration. Further, we compared features extracted from reading-a-story to counting tasks. Our results show that the number of reproducible features (out of 125) decreased stepwise with duration reduction. Spectral shape, pitch, and formants reached excellent repro-

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Reproducibility of Bio-Acoustic Features is Associated With Sample Duration, Speech Task, and Gender

Almaghrabi

Thewlis

Thwaites

et al. 2022

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…D'après les résultats déjà publiés (Henton, Bladon, 1985 ;Hanson, Chuang, 1999 ;Narra et al, 2015 ;Iseli et al, 2007 ;Klatt, Klatt, 1990) Pour le CPP, seul un effet de la langue a été observé, avec une valeur plus grande chez les hommes et femmes coréens. Nous attendions que les hommes montrent une valeur plus importante de CPP que les femmes car il existe plus de composantes harmoniques dans la voix d'hommes et leur conduit vocal est plus long que celui des femmes (Choi et Choi, 2016). Ce résultat inattendu pourrait être lié à l'influence de la tâche de production mise en évidence en coréen par Choi et Choi (2016) avec une absence d'effet en lecture.…”

Section: Résultatsunclassified

“…Cette caractéristique des plis vocaux est reflétée par d'autres paramètres acoustiques tels que le HNR (rapport harmonique/bruit) et le CPP (Cepstral Peak Prominence). Ils sont souvent utilisés pour l'évaluation de la dysphonie (Hillenbrand, Houde, 1996), mais aussi pour la comparaison de la voix en fonction de l'âge et du sexe dans les voix non-pathologiques (Chen et al, 2010 ;Choi, Choi, 2016).…”

Section: Introductionunclassified

Différences mélodiques et spectrales entre sexes comparées chez les locuteurs coréens et français

Yoon¹,

Audibert²,

Fougeron³

2022

XXXIVe Journées d'Études Sur La Parole -- JEP 2022

View full text Add to dashboard Cite

Cette étude examine la variation entre langues des différences entre sexes dans la production de la parole. Un ensemble de paramètres acoustiques liés à la fréquence fondamentale et sa variation, à la forme du spectre et au degré de périodicité sont analysés sur les productions de 70 hommes et femmes coréens et français. Les résultats montrent un effet du sexe sur la plupart des paramètres avec des valeurs plus élevées chez les femmes (f0, H1*-H2*, H1*-A1*, H1*-A2*, et HNR) excepté une valeur plus importante de H2kHz*-H5kHz chez les hommes, ce qui indique une voix perçue comme plus soufflée chez les femmes. Pour H1*-A1*, H1*-A2*, H1*-A3*, les différences entre sexes ne sont significatives que dans le groupe français. Ces résultats suggèrent qu'au-delà du dimorphisme sexuel dans les deux langues, les différences entre voix d'hommes et voix de femmes seraient moins marquées en coréen qu'en français.

show abstract

“…Hillenbrand et al(1994) (Awan & Roy, 2006;Heman-Ackah et al, 2002;Heman-Ackah et al, 2003;Kumar et al, 2010;Seo & Seong, 2013;Shin, et al, 2018;Yu et al, 2018). 특히, 모음뿐 아니라 음도변화 를 포함하는 연결발화도 분석이 가능하여 비주기적인 음성 신호 도 분석이 가능한 장점이 있다 (Choi & Choi, 2016;Heman-Ackah et al, 2003).…”

unclassified

Classification of muscle tension dysphonia (MTD) female speech and normal speech using cepstrum variables and random forest algorithm*

Yun¹,

Shim²,

Seong³

2020

Phonetics Speech Sci.

View full text Add to dashboard Cite

This study investigated the acoustic characteristics of sustained vowel /a/ and sentence utterance produced by patients with muscle tension dysphonia (MTD) using cepstrum-based acoustic variables. 36 women diagnosed with MTD and the same number of women with normal voice participated in the study and the data were recorded and measured by ADSV™. The results demonstrated that cepstral peak prominence (CPP) and CPP_F0 among all of the variables were statistically significantly lower than those of control group. When it comes to the GRBAS scale, overall severity (G) was most prominent, and roughness (R), breathiness (B), and strain (S) indices followed in order in the voice quality of MTD patients. As these characteristics increased, a statistically significant negative correlation was observed in CPP. We tried to classify MTD and control group using CPP and CPP_F0 variables. As a result of statistic modeling with a Random Forest machine learning algorithm, much higher classification accuracy (100% in training data and 83.3% in test data) was found in the sentence reading task, with CPP being proved to be playing a more crucial role in both vowel and sentence reading tasks.

show abstract

The Effect of Gender and Speech Task on Cepstral- and Spectral-Measures of Korean Normal Speakers

Cited by 19 publications

References 0 publications

The Reproducibility of Bio-Acoustic Features is Associated With Sample Duration, Speech Task, and Gender

The Reproducibility of Bio-Acoustic Features is Associated With Sample Duration, Speech Task, and Gender

Différences mélodiques et spectrales entre sexes comparées chez les locuteurs coréens et français

Classification of muscle tension dysphonia (MTD) female speech and normal speech using cepstrum variables and random forest algorithm*

Contact Info

Product

Resources

About