提出了基于小波变换理论的说话人识别算法———小波基包法 .利用小波变换多分辨的特性 ,提取出真正地反映基音周期变化的基频包络 ,从声学语音学的角度去认识方言和口音 ,首次提出了汉语口音的参量表征 ,并应用到说话人识别当中 ,以实现从听觉的角度出发 ,进行说话人的先天特性与后天特性相结合的说话人识别 .结果表明 ,从基频包络中可以表现口音的一些表征 ,有助于认识说话人的特性 . A speaker recognition algorithm based on wavelet transform theory, wavelet base-packet method, is proposed. By using the multi-resolution feature of wavelet transform, the fundamental frequency envelope, which truly reflects the pitch cycle variation, is extracted and the dialect is recognized from the perspective of acoustics phonetics And accents, the parametric characterization of Chinese accent is proposed for the first time and applied to speaker recognition in order to achieve the speaker recognition from the auditory perspective, combining the speaker’s innate and acquired features.The results show that from the fundamental frequency Some representations of accent can be expressed in the envelope, helping to understand the speaker’s personality.