摘要
Abstract
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Key words
speech enhancement/speech intelligibility/amplitude square spectrum estimation/amplitude spectral con-straint分类
信息技术与安全科学
