摘要
本文提出一个性能可靠的非特定人汉语普通话四声识别方法.该方法采用中心削波的无偏自相关法作基音周期检测,通过对基音周期进行数据选取、误差修正、平滑、拟合等处理过程,获取两维的判决矢量供四声判决.普通话单音节发音的四声识别率达98％以上.
This paper presents a reliable method for recognizing chinese speech (putonghua) tones speaker independently. This method adopts an unbiased center-clipping autocorrelation algorithm to extract pitch periods. A two-dimensional decision vector for recognizing four chinese speech tones is obtained by passing the pitch period sequence through the procedures of data selection, error correction, data smoothing and curve fitting. The average correct rate of tone recognition for isolated Chinese syllables is over 98%.
出处
《声学学报》
EI
CSCD
北大核心
1993年第5期379-385,共7页
Acta Acustica
基金
国家自然科学基金资助项目
