Title : 
Investigation of muscle activation in speech production based on an articulatory model
         
        
            Author : 
Wu, Xiyu ; Fang, Qiang ; Dang, Jianwu
         
        
            Author_Institution : 
Sch. of Inf. Sci., Japan Adv. Inst. of Sci. & Technol., Ishikawa, Japan
         
        
        
            fDate : 
Nov. 29 2010-Dec. 3 2010
         
        
        
        
            Abstract : 
In the speech acquisition process, a person starts by learning muscle activation patterns (motor commands) to control their speech organs. However, the learning process is still a brainteaser. This paper attempts to simulate this learning process using a model based method. The economy and efficiency principles are implemented in the learning process to approach the articulatory target in the somatosensory point of view. The muscle activation patterns for five Japanese vowels were obtained by the proposed learning process. The learned muscle activation patterns were consistent with the empirical data. The results were also evaluated in the articulatory and acoustic space, and these evaluations proved that the proposed method was effective.
         
        
            Keywords : 
electromyography; medical signal processing; muscle; speech processing; Japanese vowels; acoustic space; articulatory model; articulatory space; articulatory target; muscle activation investigation; somatosensory point; speech acquisition process; speech organs; speech production; Acoustics; Biological system modeling; Muscles; Principal component analysis; Production; Speech; Tongue; articulatory model; muscle activation pattern; somatosensory feedback; speech acquisition;
         
        
        
        
            Conference_Titel : 
Chinese Spoken Language Processing (ISCSLP), 2010 7th International Symposium on
         
        
            Conference_Location : 
Tainan
         
        
            Print_ISBN : 
978-1-4244-6244-5
         
        
        
            DOI : 
10.1109/ISCSLP.2010.5684876