欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2006, Vol. 23 ›› Issue (2): 174-177.DOI: 10.7523/j.issn.2095-6134.2006.2.013

• 论文 • 上一篇    下一篇

基于MPC5200的嵌入式非特定人连续语音识别系统

刘 斌; 杜利民   

  1. 中国科学院声学研究所语音交互技术中心,北京 100080
  • 收稿日期:1900-01-01 修回日期:1900-01-01 发布日期:2006-03-15

Embedded Speaker-Independent Continuous Speech Recognition System Based on MPC5200

LIU Bin, DU Li-Min   

  1. Labs for Speech Interaction Technology Research,
    Institute of Acoustics, Chinese Academy of Sciences, Beijing 100080
  • Received:1900-01-01 Revised:1900-01-01 Published:2006-03-15

摘要: 非特定人连续语音识别系统具有非常高的计算复杂度,对系统的硬件资源有很高的要求,本文针对该问题设计了一款基于MPC5200微处理器的嵌入式平台,在该平台上实现了基于隐含马尔可夫模型的非特定人连续语音识别系统,实验结果表明该嵌入式语音识别系统达到了实时识别的要求。

关键词: 语音识别, 隐含马尔可夫模型, 嵌入式系统, MPC5200

Abstract: The embedded systems’ ability is not enough to run HMM-based speaker-independent continuous speech recognition algorithms, especially the search algorithm and the state likelihood computation. This paper proposes an embedded platform based on MPC5200 microprocessor, and implements a HMM-based speaker-independent continuous speech recognition system on this embedded platform. The experiments show that the real-time performance of this embedded speech recognition system is acceptable.

Key words: speech recognition, HMM, embedded system, MPC5200

中图分类号: