Removal of Empty Bands from Speech Signals in Hindi

In order to provide speech analysis, speech production and perception can be exploited with the help of various speech processing applications. Few properties or features can be extracted from speech signal S(n) with the help of speech analysis. With the objective of simplifying the speech signal and for removing the redundancy present within the speech signal, the transformation of S(n) is done into another signal. In this research work, the novel technique is proposed to remove empty bands from the speech signal. The proposed technique is based on the threshold value for the removal of empty bands. In the proposed technique, the frequency of each band is calculated and a band which has frequency below threshold value is removed from the signal. The performance of the proposed technique is tested in MATLAB and it is analyzed that proposed technique performs well in terms speech enhancement.

 Key words: Empty bands, LDA.