一种基于幅度谱约束的语音可懂度增强算法  

A Speech Intelligibility Enhancement Algorithm Based on Amplitude Spectrum Constraint

作　　者：赵艳磊 辛凯 ZHAO Yanlei;XIN Kai(College of Jiangsu Aviation Vocational and Technical,Zhenjiang 212003;School of Computer Science and Communication Engineering,Jiangsu University,Zhenjiang 212003)

机构地区：[1]江苏航空职业技术学院,镇江212003 [2]江苏大学计算机科学与通信工程学院,镇江212003

出　　处：《计算机与数字工程》2025年第1期39-44,共6页Computer & Digital Engineering

摘　　要：语音交流是重要的人机交互方式,然而现实生活中噪声干扰影响接收到的语音信号综合质量。因此对受到噪声污染的带噪语音进行处理,以尽可能地恢复出原始纯净语音信号、保证语音通信效果。论文围绕基于语音幅度谱约束的语音可懂度增强问题开展了研究和分析。首先,给出语音幅度谱Ⅰ+Ⅱ约束算法的推导过程以及该算法提升语音可懂度的设计思路。其次,围绕Ⅰ+Ⅱ约束算法需要纯净语音信号的先验信息,无法在实际中得到应用的问题,将语音幅度平方谱估计理论与Ⅰ+Ⅱ约束算法相融合提出了两个可实际应用的语音可懂度增强算法。最后,采用多种噪声环境及语音和噪声数据对提出算法的性能进行了仿真验证,有效说明了提出算法在提升带噪语音可懂度等方面的优良性能。Speech communication is a momentous way of human-computer interaction.However,noise interference in real life affects the overall quality of the received voice signal.Therefore,it is necessary to use the speech enhancement technology to process the speech contaminated by noise,restore the original signal as accurately as possible,and improve the speech communica⁃tion quality.In this paper,the problem of speech intelligibility enhancement based on speech amplitude spectrum constraint is stud⁃ied and analyzed.Firstly,the speech amplitude spectrum of I+II constraint algorithm's derivation process is given,the basic princi⁃ples of the algorithm and its design ideas for improving speech intelligibility are explained.Secondly,around the problem that the al⁃gorithm requires the prior information of the clean speech signal and cannot be applied in practical.Combined the speech amplitude square spectrum estimation theory with the I+II constraint algorithm,based on the complex Gaussian hypothesis model of speech and noise,two practically applicable speech intelligibility enhancement algorithms are proposed according to the maximum posterior and minimum mean square error methods.Finally,the performance of the proposed algorithm is simulated and verified by using a va⁃riety of noise environments and the data of speech and noise.It effectively shows that the proposed algorithm can significantly im⁃prove the performance of intelligibility of noisy speech.

关 键 词：语音增强 语音可懂度 幅度平方谱估计 幅度谱约束 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]