语音识别中DTW模型的改进算法研究

Research on the Improved Algorithm of DTW Model in Speech Recognition

对基于 DTW 算法的语音识别数学模型作了深入研究,针对传统 DTW 算法在实际应用时诸多缺点提出改进算法(改进端点检测,改进搜索路径、松弛起终点,设置阈值),用 MATLAB 对改进算法进行了仿真,并对仿真结果进行分析与比较,得到了很好的验证,提高了语音识别率,节省了存储空间。

Aiming at the disadvantages of the traditional DTW algorithm in practical application, the mathematical model of speech-recognition based on DTW algorithm is researched intensively and some improved algorithms are put forward, such as the improved end-point detection, the im- proved searching path, the slacked of start-point and end-point, setting the limited match-value and so on. The improved algorithms are carried out simulation with MATLAB, and the results are analyzed and compared. The rationalization of the proposed algorithms is validated through simulation. The algorithms enhance speech recognition rate and save memory-space.