Abstract:There is a big challenge in speech recognition system due to variability of the spoken languages and also speech signal is degraded due the environmental noise. Therefore speech recognition system requires pre-processing, which plays vital role to restore speech signal more effectively. Many challenges are there to restore the speech from different noisy environment. In this paper, proposed method is divided into two parts. First part is Kannada speech restoration, the modified wiener filter is proposed to restore the Kannada speech signal with very good speech quality. The experimental results are verified for different types of noise by varying input Signal to Noise Ratio (SNR) from -20dB to 5 decibels (dB). Second part is Autocorrelation based Kannada speech recognition system. Autocorrelation technique gives better output SNR for degraded input SNR with -20dB. The Autocorrelation is the simple method to recognize the isolated Kannada word in speech signal. This proposed technique results 100% recognition rate for male, female, with different accents.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.