论文部分内容阅读
随着嵌入式人机交互系统的发展,语音识别技术成为其实现的重要途径。本文介绍了一种基于SPCE061A16位单片机的嵌入式语音识别系统,着重分析了上升过零率和峰值幅度(ZCPA)特征提取算法和增强型可变速率编码(EVRC)除噪算法等预处理技术,基于隐马尔可夫(HMM)模型的该系统可以实现语音识别的功能,可以为开发综合功能更强大的人机自然交互系统提供帮助与支持。
With the development of embedded human-computer interaction system, speech recognition technology has become an important way to realize it. This paper introduces an embedded speech recognition system based on SPCE061A16 singlechip microcomputer. The pretreatment techniques such as the rising zero-crossing rate and peak amplitude (ZCPA) feature extraction algorithm and the enhanced variable rate coding (EVRC) The system based on Hidden Markov Model (HMM) can realize the function of speech recognition, which can help and develop the more powerful man-machine interaction system with integrated functions.