【摘要】 研究语音特征参数准确提取的方法。利用传统算法进行语音特征参数提取,需要获取语音特征的连续谱,根据该连续谱提取其特征参数。但是由于语音特征的随机性较强,很难保证其连续性特征,因此将造成语音特征参数提取的准确性降低。为了避免上述缺陷,提出了一种基于谐波探测算法的语音特征参数提取方法。详细阐述谐波探测算法的相关理论,将该理论应用于语音特征参数提取的过程中。利用该算法进行语音特征参数提取。实验结果表明,利用本文算法进行语音特征参数提取优化处理,能够极大的提高语音特征参数提取的准确性,取得了令人满意的结果。更多还原

【Abstract】 The speech feature parameter extraction method accurately. Using traditional algorithm speech feature parameter extraction,we need to get the phonetic characteristics of continuous spectrum,according to the continuous spectrum to extract the feature parameters. But because of the phonetic characteristics of strong randomness,it is dif- ficult to guarantee the continuity characteristics,thus will cause decrease the accuracy of speech feature parameter extraction. In order to avoid these defects,a speech feature parameter extraction method based on harmonic detection algorithm. In detail in this paper,the relevant theories of harmonic detection algorithm,applies the theory of speech feature parameter extraction process. Using the algorithm for speech feature parameter extraction. The experimental results show that the algorithm presented in this paper to carry on the speech feature parameter extraction and optimi- zation of processing,can greatly improve the accuracy of speech feature parameter extraction,achieved satisfactory results.更多还原