首页> 中文期刊>电子设计工程 >一种基于模板匹配的语音识别算法

一种基于模板匹配的语音识别算法

     

摘要

Speech recognition is an important research direction of Speech signal processing, which is related to physiology, psychology, linguistics, computer science and signal processing and many other fields. It is widely used in control, communications, digital products and other industries. This article describes a simple speech recognition algorithm. The main processes include pre-processing, endpoint detection, extraction of feature value and pattern matching. The amplitude is the standard in endpoint detection. The critical Eigen Value is used as eigenvector. DTW algorithm is used in pattern matching. Main design idea is to get the voice signal and match it with the stored pattern. A hardware system with TMS32OVCS402 chip as the core is used to implement this algorithm.%语音识别是语音信号处理的一个重要研究方向,涉及到生理学、心理学、语言学、计算机科学以及信号处理等诸多领域,广泛应用于控制、通信、消费品等行业。文中介绍了一种简单的语音识别算法。该算法主要流程包括预处理、端点检测、提取特征值、模式匹配4个过程,其中端点检测以幅值为标准,特征值采用临界带特征矢量,模板匹配采用DTW算法。主要设计思想是将得到的语音信号和已经存储的模板进行匹配。实践上用TMS320VC5402芯片为核心的系统硬件来实现这个算法。

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号