Abstract
ABSTRACTOccurrence of auditory masking is a consequence of close frequency components or grouping of same frequencies of multiple signals. Earlier studies reported that the effect of increased spectral masking is reduced by using binaural dichotic appearance using critical bandwidth-based spectral splitting with comb filters. In the proposed work, spectral splitting of speech signals is accomplished by using optimised wavelet packets with Symlet and Daubechies family. Here, eight bands possessing the bandwidth known as quasi octave bandwidth were created. Seven participants with sensorineural hearing loss had been considered and evaluated through experimentation. Further to randomise the tests, 15 syllables without meaning were considered in vowel consonant vowel (VCV) order. Results indicated a significant decrease in response time and improvement in recognition score, speech quality and transmission of overall features over the unprocessed signal.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
