Unvoiced Speech Recognition Based on One-Channel Facial Myoelectric Signal

Six Chinese vowels /a/, /o/, /e/, /i/, /u/, and /u/ are recognized based on the one-channel detected facial myoelectric signal (MES). Zygomaticus major and anterior belly of the digastric are carefully selected as the electrodes sites of MES detected. Over-sampling technology and four-layer wavelet decomposition are used to reduce noise in MES records. Digital down converter down converts the original sampling rate to Nyquist frequency. By anterior methods, clean MES is gotten with no signal distortion. For MES is not voice signal, MFCC is not selected as the feature set. According to MES characteristic, ten-order AR model is set up. The coefficients of AR model, cepstral coefficients, and amplitude of MES are chosen to form the original feature set for recognition. Principal components analysis (PCA) reduces the dimension of original feature set before the proposed BP networks. Combining two BP network classifiers, an efficient classifier is proposed. The result of experiment shows that Chinese vowel /u/, and /i/ have good classification rate (more than 90%) based on one-channel facial myoelectric signal

[1]  George N. Saridis,et al.  EMG Pattern Analysis and Classification for a Prosthetic Arm , 1982, IEEE Transactions on Biomedical Engineering.

[2]  Adrian D. C. Chan,et al.  Continuous myoelectric control for powered prostheses using hidden Markov models , 2005, IEEE Transactions on Biomedical Engineering.

[3]  M. S. Morse,et al.  A proposed control scheme for a vocal prosthesis , 1990, IEEE Proceedings on Southeastcon.

[4]  Kevin Englehart,et al.  A multi-expert speech recognition system using acoustic and myoelectric signals , 2002, Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society] [Engineering in Medicine and Biology.

[5]  H. Manabe,et al.  Multi-stream HMM for EMG-based speech recognition , 2004, The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[6]  Toshiaki Sugimura,et al.  "Unvoiced speech recognition using EMG - mime speech recognition" , 2003, CHI Extended Abstracts.

[7]  N. Bu,et al.  Phoneme Classification for Speech Synthesiser using Differential EMG Signals between Muscles , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[8]  M. Pinsky.  Introduction to Fourier analysis and wavelets , 2002 .

[9]  Todd A. Kuiken,et al.  Independence of myoelectric control signals examined using a surface EMG model , 2003, IEEE Transactions on Biomedical Engineering.

[10]  H. Manabe,et al.  Robust and Preceding Speech Detection Using EMG , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[11]  M.S. Morse,et al.  Use of myoelectric signals to recognize speech , 1989, Images of the Twenty-First Century. Proceedings of the Annual International Engineering in Medicine and Biology Society,.

[12]  K. Nazarpour,et al.  Surface EMG Signal Classification Using a Selective Mix of Higher Order Statistics , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.