Effect of Spectrally Modified Speech on Speech Perception of Native Speakers of Marathi

Abstract

Objective: We investigated the effect of spectrally modified speech on speech perception of native speakers of Marathi. Methods: A total of eight word-lists with each list consisting of 25 words in Marathi were used as speech stimuli. The speech stimuli in List 1, List 2, List 3, List 4, List 5, List 6, and List 7 were spectrally modified by filtering them at 500 Hz, 1000 Hz, 1500 Hz, 2000 Hz, 2500 Hz, 3000 Hz, and 3500 Hz cutoff frequencies, respectively. The speech stimuli in List 8 were unfiltered. The speech identification score (SIS) testing was performed as a measure of speech perception using eight word-lists on a total of 125 normal hearing native speakers of Marathi in the age range between 18 and 25 years. Results: The results revealed that there was a statistically significant effect of spectral modification on speech perception. An increase in cutoff frequency of speech stimuli was analogous to increase in speech perception performance. The normal range of SIS (90%–100%) was achieved for speech stimuli filtered at 2500 Hz cutoff frequency. Conclusion: Although speech is a broadband signal, the findings of this study suggest that the spectral information till 2500 Hz is sufficient for optimum perception of speech of Marathi.