The Design and Implementation of Hybrid Engine used in English interaction learning system

With the continuous improvement and development of Hybrid recognition technology, the numerous special purpose chips for Hybrid recognition have been developed, thus, the practical products of Hybrid recognition have been gradually appeared in the market. This paper will take the foundation of Hybrid coding in English embedded Hybrid system as the breakthrough point, by means of the interpretation of Speex, it discusses the design of the Hybrid engine as well as the route of implementation so as to reduce tedious work in voice testing. Combined with the analysis of the basic structure of Hybrid recognition in the embedded system, it discusses the architecture of the speaker independent English Hybrid recognition learning system.

[1]  Mark J. F. Gales,et al.  Factor analysed hidden Markov models , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Les E. Atlas,et al.  Applications of positive time-frequency distributions to speech processing , 1994, IEEE Trans. Speech Audio Process..

[3]  Curtis P. Langlotz,et al.  Improving language models for radiology speech recognition , 2009, J. Biomed. Informatics.