藏语语音的清浊音识别

Tibetan Speech of Voiceless and Voiced Recogenition

清浊音识别在语音处理技术中具有非常重要的意义,其准确度对后续的语音处理有很大的影响.文章根据藏语言文字的特征提取了藏语有声段语音的短时能量、短时过零率以及短时自相关等参数,并通过对语音进行短时时域分析,研究了藏语的清浊音识别方法.实验证明,此方法在藏语辅音字母的清浊音识别中获得良好的效果,该方法为进一步深入研究藏语语音技术提供了一定的参考依据.

The voiceless and voiced recognition is very important in speech processing technology,and its accuracy has a great influence on subsequent speech processing.In this paper,the short-time energy,short-time zero crossing rate and the short-time autocorrelation of Tibetan vocal segments are extracted according to the characteristics Tibetan language and through a short time domain analysis of the speech,to study the voiceless and voiced recognition method for Tibetan.The experiment proves that the method of this paper obtains good results in the recognition of the voiceless and voiced of Tibetan consonant letters,which provided some experience and foundation for further study of Tibetan speech technology.