COMPENSATION OF SPEAKER DIRECTIVITY IN SPEEC USING HMM COMPOSITION

efficiency of this method when compared to speech acquisi-. In hands-free speech recognition the speaker should be able to move freely in front of the speech acquisiti'on device. However, the speech signal is then submitted to variations due to the continuous change of position in the acoustic space. This paper focuses on the role of speaker head rotations as compared with static situations in anechoic conditions. The effect of speaker directivity in speech recognition performance degradation is demonstrated and a compensation method based on HMM composition is proposed to increase the performance.

[1]  Katsuhiko Shirai,et al.  Japanese large-vocabulary continuous-speech recognition using a business-newspaper corpus , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Sadaoki Furui,et al.  Adaptation method based on HMM composition and EM algorithm , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[3]  Sofiène Affes,et al.  Microphone array response to speaker movements , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[4]  Satoshi Nakamura,et al.  Noise and room acoustics distorted speech recognition by HMM composition , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.