Advancement Is Independent Component Analysis Speech Enhancement Process

Abstract

Speech enhancement is a process of increasing the quality, intelligibility of the speech signal which is one dimensional. So, by removing the external world noise or artefacts this can be improve and reject the background interference in the form of additive background noise which helps in the improvement of performance of communication system. The enhancement of speech signal consists of 3 major objectives: A) Sound is to be clear to the listener. B) Improve robustness. C) Lossless perception is required which improves the accuracy. For loss enhancement subspace methods were used such as Spectral Subtraction method, Time domain method, Transform method. The signal enhancement process is going to be perform by using sub band decomposition method under low and high pass filter collaboration i.e., WAVELET (DWT) transformation. The existing technique KLT sub space is a loss transformation and performed enhancing signal but theoretically it can be improved by Wavelet transformation and will perform a lossless transformation technique.