Speech enhancement by changing complex spectrum compensation
Priyanka Kumari, Kratika Sharma
In speech processing, enhancement of speech signal can be defend by multiple enhancement techniques. Background noise is one of the common problem in speech processing, due to this the quality and accuracy of speech reduces automatically. Previously, many of the speech enhancement algorithms works only on short- time magnitude spectrum, while keeping short- time magnitude spectrum remains unchanged or else operates only on short- time phase spectrum by keeping short- time magnitude spectrum remains the same. There is no such technique was implemented to work on both the spectrums to enhance the quality of the speech signal. In this paper, a novel speech enhancement technique is proposed to change the characteristics of both the magnitude and phase spectrums to produce a modified complex spectrum with improved speech quality. The test of an objective speech quality measure PESQ, and spectrogram analysis had showed that the proposed method can obtain better enhancement performance.