DOI: 10.21276/sjet.2017.5.9.6

Pages: 488-492

Speech enhancement aims to improve speech quality by using various algorithms. The objective of enhancement is improvement in intelligibility using audio signal processing techniques. The choice of the filter-bank has a significant influence on the performance of such systems in terms of signal quality, computational complexity and signal delay. Accordingly, the filter-bank design has to fulfill different, partly conflicting requirements in dependence of the considered application. In this paper we tried to design Low Pass, High pass & band pass Filters to process Natural Spoken Marathi alphabets. We also processed Segmented & overlapped Alphabets to compare our algorithm performance. Major perspective to design filter bank is to select particular filter band to process respective vowel & consonant.