DocumentCode :
735035
Title :
Automatic evaluation of resonance and articulation disorders in cleft palate speech
Author :
Ling He ; Jie Tan ; HuaQing Hao ; Ming Tang ; Heng Yin ; Lech, Margaret
Author_Institution :
Sch. of Electr. Eng. & Inf., Sichuan Univ., Chengdu, China
fYear :
2015
fDate :
12-15 July 2015
Firstpage :
358
Lastpage :
362
Abstract :
The evaluation of cleft palate (CP) speech is a critical clinical treatment. The most typical characteristics of CP speech include hypernasality and consonant misarticulation. Currently, the evaluation of CP speech is carried out by experienced speech therapists. It strongly depends on their clinical experience and subjective judgment. This work aims to propose an automatic evaluation system of resonance and articulation disorders in CP speech. The CP speech database is collected by the Hospital of Stomatology, Sichuan University, which has the largest number of CP patients in China. The automatic hypernasality grading algorithm is proposed to classify four levels of hypernasality: normal, mild, moderate and severe. Besides, the algorithms of automatic consonant omission and replacement location are proposed to evaluate the speech intelligibility of CP speech.
Keywords :
medical disorders; patient treatment; speech intelligibility; speech processing; CP speech database; Hospital of Stomatology; Sichuan University; articulation disorders; automatic hypernasality grading algorithm; cleft palate speech; clinical treatment; consonant misarticulation; evaluation system; speech intelligibility; speech therapist; typical characteristics; Accuracy; Acoustics; Cavity resonators; Databases; Feature extraction; Speech; Speech processing; Cleft palate speech; articulation disorder; consonant misarticulation; hypernasality; resonance disorder;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal and Information Processing (ChinaSIP), 2015 IEEE China Summit and International Conference on
Conference_Location :
Chengdu
Type :
conf
DOI :
10.1109/ChinaSIP.2015.7230424
Filename :
7230424
Link To Document :
بازگشت