Research and improvement on embedded system application of DTW-based speech recognition

Dynamic time warping (DTW) algorithm is one of the most popular mathematical models in the field of speech recognition. The algorithm is simple and effective. It is almost the same as HMM algorithm on speech recognition score and other characteristics, and it is widely applied to some special fields such as resource poor embedded system for its strong point of simple and effective algorithm. The paper describes an improved DTW-based speech recognition model, and shows a DTW-based application of portable value-added tax calculator with speaker-dependent connected-word and isolated-word speech recognition abilities built in.

[1]  Sun Cheng-li.  On chip realization of speaker-independent isolated word speech recognizer , 2007 .

[2]  Li Deng.  Integrated optimization of dynamic feature parameters for hidden Markov modeling of speech , 1994, IEEE Signal Processing Letters.

[3]  John H. L. Hansen,et al.  Discrete-Time Processing of Speech Signals , 1993 .

[4]  Liu Qingsheng.  Research on a Speech Endpoint Detection Method , 2003 .

[5]  Li Deng.  Integrated optimization of dynamic feature parameters for hidden Markov modeling of speech , 1994, IEEE Signal Process. Lett..

[6]  Wang Jinjun.  Application of Speaker-independent Speech Recognition in Embedded Systems , 2006 .

[7]  Ahmet M. Kondoz,et al.  Digital Speech: Coding for Low Bit Rate Communication Systems , 1995 .

[8]  Ronald W. Schafer,et al.  Digital Processing of Speech Signals , 1978 .

[9]  Monica N. Nicolescu,et al.  RECOGNIZING SIMPLE HUMAN ACTIONS USING 3D HEAD MOVEMENT , 2007, Comput. Intell..

[10]  Akira Hayashi,et al.  Embedding of time series data by using dynamic time warping distances , 2006 .

[11]  Seiichi Nakagawa,et al.  Detection and recognition of correction utterances on misrecognition of spoken dialog system , 2005 .

[12]  M.G. Bellanger,et al.  Digital processing of speech signals , 1980, Proceedings of the IEEE.

[13]  Kim-Fung Man,et al.  Genetic Time Warping for Isolated Word Recognition , 1996, Int. J. Pattern Recognit. Artif. Intell..

[14]  Sumit Bose,et al.  An Approach to Proper Speech Segmentation for Quality Improvement in Concatenative Text-To-Speech System for Indian Languages , 2005, Int. J. Comput. Process. Orient. Lang..