Detection and assessment of hypernasality in repaired cleft palate speech using vocal tract and residual features

2019 ◽  
Vol 146 (6) ◽  
pp. 4211-4223
Author(s):  
Akhilesh Kumar Dubey ◽  
S. R. Mahadeva Prasanna ◽  
S. Dandapat
Keyword(s):  
Cleft Palate ◽  
Vocal Tract ◽  
Cleft Palate Speech

Automatic hypernasality grade assessment in cleft palate speech based on the spectral envelope method

2020 ◽  
Vol 65 (1) ◽  
pp. 73-86 ◽  
Author(s):  
Jing Zhang ◽  
Sen Yang ◽  
Xiyue Wang ◽  
Ming Tang ◽  
Heng Yin ◽  
...  
Keyword(s):  
Cleft Palate ◽  
Speech Intelligibility ◽  
Linear Prediction ◽  
Vocal Tract ◽  
Experimental Results ◽  
Spectral Envelope ◽  
Support Vector ◽  
Cleft Palate Speech ◽  
Hypernasal Speech ◽  
Normal Speech

AbstractDue to velopharyngeal incompetence, airflow overflows from the oral cavity to the nasal cavity, which results in hypernasality. Hypernasality greatly reduces speech intelligibility and affects the daily communication of patients with cleft palate. Accurate assessment of hypernasality grades can provide assisted diagnosis for speech-language pathologists (SLPs) in clinical settings. Utilizing a support vector machine (SVM), this paper classifies speech recordings into four grades (normal, mild, moderate and severe hypernasality) based on vocal tract characteristics. Linear prediction (LP) analysis is widely used to model the vocal tract. Glottal source information may be included in the LP-based spectrum. The stabilized weighted linear prediction (SWLP) method, which imposes the temporal weights on the closed-phase interval of the glottal cycle, is a more robust approach for modeling the vocal tract. The extended weighted linear prediction (XLP) method weights each lagged speech signal separately, which achieves a finer time scale on the spectral envelope than the SWLP method. Tested speech recordings were collected from 60 subjects with cleft palate and 20 control subjects, and included a total of 4640 Mandarin syllables. The experimental results showed that the spectral envelope of normal speech decreases faster than that of hypernasal speech in the high-frequency part. The experimental results also indicate that the SWLP- and XLP-based methods have smaller correlation coefficients between normal and hypernasal speech than the LP method. Thus, the SWLP and XLP methods have better ability to distinguish hypernasal from normal speech than the LP method. The classification accuracies of the four hypernasality grades using the SWLP and XLP methods range from 83.86% to 97.47%. The selection of the model order and the size of the weight function are also discussed in this paper.

A Practical Method of Judging Cleft Palate Speech*

1966 ◽  
Vol 38 (6) ◽  
pp. 544-551
Author(s):  
JEROME E. ADAMSON ◽  
CHARLES E. HORTON ◽  
HUGH H. CRAWFORD ◽  
MILDRED WERTZ
Keyword(s):  
Cleft Palate ◽  
Practical Method ◽  
Cleft Palate Speech

The Correlation between Nasalance and a Differentiated Perceptual Rating of Speech in Dutch Patients with Velopharyngeal Insufficiency

2002 ◽  
Vol 39 (3) ◽  
pp. 277-284 ◽  
Author(s):  
Kornelis H.D.M. Keuning ◽  
George H. Wieneke ◽  
Hans A. Van Wijngaarden ◽  
Philippe H. Dejonckere
Keyword(s):  
Cleft Palate ◽  
Correlation Coefficients ◽  
University Hospital ◽  
Velopharyngeal Insufficiency ◽  
Visual Analog Scales ◽  
Cleft Palate Speech ◽  
Normal Text ◽  
The Mean ◽  
Lower Correlation ◽  
Nasal Emission

Objective: The correlation between the nasalance score and the perceptual rating of several aspects of speech of speakers with velopharyngeal insufficiency (VPI) by six speech-language pathologists was evaluated. Procedure: The overall grade of severity, hypernasality, audible nasal emission, misarticulations, and intelligibility were rated on visual analog scales. Speech samples with a normal distribution of phonemes (normal text [NT]) and those free of nasal consonants (denasal text [DT]) of 43 patients with VPI were used. Mean nasalance scores were computed for the speech samples, and Spearman correlation coefficients were computed between the mean nasalance score and the five parameters of the differentiated rating. Setting: The Institute of Phoniatrics, Utrecht University Hospital, The Netherlands. Results: The correlation coefficient between the mean nasalance and the perceptual rating of hypernasality ranged among judges from .31 to .56 for NT speech samples and .36 to .60 for DT speech samples. Only small differences were found between speech pathologists with and without expertise in cleft palate speech. The rating of the overall grade of severity appeared to correlate quite well with the rating of the intelligibility (rNT = .77, rDT = .79). Lower correlation coefficients, ranging from .34 to .71, were found between overall grade of severity and hypernasality, audible nasal emission, and mis-articulations. Conclusions: A low correlation between the nasalance and the perceptual rating of hypernasality was found. The parameter overall grade of severity appeared to be determined mainly by the parameter intelligibility. Expertise in rating of cleft palate speech does not guarantee a high correlation between instrumental measurement and perceptual rating.

scholarly journals Automatic detection of pharyngeal fricative in cleft palate speech

2018 ◽  
Vol 246 ◽  
pp. 03007
Author(s):  
Fei He ◽  
Geyi Zhou ◽  
Xinyi He ◽  
Heng Yin ◽  
Ling He
Keyword(s):  
Cleft Palate ◽  
Nearest Neighbor ◽  
Automatic Detection ◽  
Acoustic Parameter ◽  
Detection Accuracy ◽  
K Nearest Neighbor ◽  
Acoustic Feature ◽  
Nearest Neighbor Classifier ◽  
Cleft Palate Speech ◽  
Neighbor Classifier

Pharyngeal fricative occurs during the production of consonants, which makes the consonants lose or weaken in cleft palate speech. In clinical application, the automatic detection of pharyngeal fricative in cleft palate speech could provide objective and effective assistant aids for speech language pathologists. In this paper, a novel acoustic parameter is proposed to detect the existence of pharyngeal fricative in cleft palate speech. This proposed acoustic feature ICPD (Independent Consonant Prominent Distribution) reflects the movement of mouth and tongue. The experimental results show that normal fricative has the higher ICPD. The extracted ICPD feature is combined with k-nearest neighbor classifier to achieve the automatic detection of pharyngeal fricative. The proposed system is tested on 127 speech samples recorded by cleft palate patients and 94 by normal speakers of controls. The overall pharyngeal fricative detection accuracy is around 90%.

Sign in / Sign up

Export Citation Format

Share Document