摘要
本文主要论述了一种小词表语音识别系统的硬、软件设计方法。系统以DSP5416为硬件平台，采用非线性美尔刻度倒谱参数（MFCC）特征参数提取算法，动态时间规整（DTW）作为识别算法，实现了语音识别系统的设计。实验结果表明平均语音识别率不低于90％，取得良好的识别效果。
The thesis mainly pays attention to the design of software and hardware circuits about the small glossary. The system takes DSP as hardware platform, adopting the MFCC as a characteristic parameter drawing algorithm, and taking the DTW as the distinguishing algo- rithm, having realized Speech recognition system＇s. The experiment bear fruit indicates that the average of voice recognition being lower than 90%, getting fine identification effect.
作者
王桂珍
赵明建
WANG Gui-zhen,ZHAO Ming-jian （Dept, of Computer Engineering, Nanjing Institute of Technology, Nanjing 211167, China）
出处
《电脑知识与技术》
2008年第5期711-712,719,共3页
Computer Knowledge and Technology
