小词表实时语音识别系统的定点DSP实现  

Fixed Point DSP Application of A Small Glossary Real Time Speech Recognition System

作　　者：曾日波[1] 

机构地区：[1]华中科技大学控制科学与工程系,湖北武汉430074

出　　处：《现代电子技术》2004年第11期62-64,共3页Modern Electronics Technique

摘　　要：主要介绍一种基于定点数字信号处理器 DSP芯片 ( TMS3 2 0 V C5 40 2 )的小词表实时语音识别系统。该系统体积小 ,成本低 ,可扩展性好 ,方便应用于许多特定场合 ,如 :门禁控制系统等 ,对于 10 0词以内的特定人语音 ,识别的准确率超过 95 %。This article introduces a small glossary real time speech recognition system which is based on the fixed point digital signal processor (DSP) chip. This system has small scale, low cost and high capability of expanding. It is very convenient for some special situations. Such as the gating system,with in 100 glossaries of the specific person′s pronunciation, the recognition accuracy of this system is over 95%.

关 键 词：语音识别 (DSP)定点运算 线性预测编码(LPC) ADPCM算法 动态时间归整(DTW) 

分 类 号：TN912.34[电子电信—通信与信息系统]