DocumentCode :
3173213
Title :
A Preliminary Study on Emotions of Chinese Whispered Speech
Author :
Chenghui, Gong ; Heming, Zhao ; Wei, Zou ; Yanlei, Wang ; Min, Wang
Author_Institution :
Sch. of Phys. Sci. & Technol., Soochow Univ., Suzhou, China
Volume :
2
fYear :
2009
fDate :
25-27 Dec. 2009
Firstpage :
429
Lastpage :
433
Abstract :
This paper focuses on the emotional Chinese whispered speech. The auditory perception is to demonstrate that whispered speech can also carry emotional information as the voiced one do. The experiment carried on the single tonal whispered Mandarin shows that the falling-raising and falling tones could transfer feelings better than the high-level and raising ones. The accuracy of perception indicates that anger, sad and afraid are more easily to be perceived. The statistic of misperception in this test illustrates the similarity between each emotion. Apart from the perception experiment, the acoustic analysis is also conducted. The prosodic features, including short time energy and speak rate, voice quality parameter-formant, and spectrum are abstracted to analyze the differences between emotions of Chinese whispered speech. Meanwhile, the feasibility of emotional identification of whispered speech is certified.
Keywords :
acoustics; emotion recognition; hearing; natural language processing; speech processing; Chinese whispered speech; acoustic analysis; auditory perception; emotion identification; emotional information; emotional recognition; single tonal whispered Mandarin; voice quality parameter; Acoustic testing; Application software; Computer applications; Emotion recognition; Frequency; Loudspeakers; Physics computing; Signal to noise ratio; Speech analysis; Speech recognition; acoustic feature; auditory perception; formant abstraction; spectrum analysis; whispered speech;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science-Technology and Applications, 2009. IFCSTA '09. International Forum on
Conference_Location :
Chongqing
Print_ISBN :
978-0-7695-3930-0
Electronic_ISBN :
978-1-4244-5423-5
Type :
conf
DOI :
10.1109/IFCSTA.2009.227
Filename :
5384654
Link To Document :
بازگشت