基于语音压缩感知观测序列非重构的清浊音判别法.pdfVIP

Vo l. 28 No.3 第28卷第3期 数据采集与处理 2013 年 5 月 Journal of Data Acquisition and Processing May 2013 文章编号:1004-9037(2013)03-0274-06 基于语音压缩感知观测序列非重构的清浊音判别法 王文娟1.2 杨震1 ， 2 (1.南京邮电大学通信与信息工程学院，南京，210003; 2. 南京邮电大学宽带无线通信与传感技术教育部重点实验室，南京，210003) 摘要:针对语音产生的数字模型来分析清音和浊音的特点，研究了清音和浊音的压缩感知观测1 序列的特性，从而 推出清音的压缩感知观测序列具有近似高斯信号的特性，而浊音的压缩感知观测序列具有非高斯信号的特性。 基于这种特性来设计一种直接针对压缩感知观测序列、非重构的清浊音判别方法，并给出了它与重构信号清浊 音判别在准确性和计算量两个方面的比较结果，解决了基于语音压缩感知观测序列重构情况下判别清浊音的高 计算量问题。 关键词:压缩感知;清浊音判别算法;观测序列;三阶累积量 中图分类号:TN912.3 文献标志码:A Voicing-State Identification ßased on Speech Observation Sequence and Non-reconstruction in Compressing Sensing Wang WeηJuaη 1. 2 ， Yang Zhen 1. 2 (1. College of Telecommunication Information Engineering , Nanjing University of Posts and Telecommunications , Nanjing ,210003 , China; 2. Key Lab of Broadband Wireless Communication and Sensor Network Technology , Nanjing University of Posts and Telecommunications , Nanjing , 210003 , China) Abstract: Based on the theory of compressed sensing , the observation sequence after compress- ing sensing is different from the Nyquist sequence , so the voicing-state identification can be a- chieved only by reconstructing the original speech signal with high complexity. The voicing- state characteristics are analyzed based on the speech digital model , and a conclusion can be drawn that the unvoiced observation sequence has the characteristics of Gaussian sign