Research Of Voice Control Instructions Recognition Arithmetic In Vehicle

With the development of automotive technology, more and more control technology is applied to car to improve the safety, ride-comfort and entertainment. During the recent years, voice recognition technology has made remarkable progress, and voice products begin from the laboratory to the market. Meanwhile, the technology is paid more attention to apply to automotive system to implement the intelligent control to the partial electronic & electric equipments. An improved algorithm DTW is used in this paper to realize the software part of vehicular speech recognition. It is non-specific, small vocabulary, isolated word speech recognition system.At first, the present primary speech recognition methods are generally introduced. The form and the principle of typical speech recognition system are systematical illuminated in the paper. According to the feature of speech recognition system which is designed in the paper, DTW speech recognition method is determined as the recognition method of this system compared to other speech recognition methods, and the principle of DTW arithmetic is generally discussed, searching process of matching path is particularly analyzed, the estimation rule of speech recognition result is given. In allusion to the disadvantage of DTW'vast calculation and data storage, a new modified DTW arithmetic using subsection warping and searching the definite parallelogram domain is carried out to implement the model matching to adapt to the demand of system hardware resource. In the software algorithm, SAE-SAZ(short-time average energy and short-time average zero-crossing rate) is adopted in the paper to realize the endpoint. LPC coefficient of 12 bands and 12 bands LPCC coefficient of the common characteristics of the 24-dimensional vector is used as a voice signal feature vector. Next, MATLAB software is applied to simulate the voice signal processing process in order to prove the arithmetic is right.In order to inspect the recognition effect of vehicular speech recognition system, the speech recognition experiment is designed to carry on the test of system performance in the different background noise environment, and at last the experiment got a considerable recognition effect and meet the design requirements.