Musical Noise Removal of Enhanced Speech Using Gray Level Values of Spectrogram Plots 137 Musical Noise Removal of Enhanced Speech Using Gray Level Values of Spectrogram Plots

Enhanced speech using Power Spectral Subtrac-tion method is computationally quite simple but re-sults in an annoying residual noise also called musi-cal noise due to its narrowband spectrum and pres-ence of tone like characteristics. To eliminate this modified spectral subtraction technique is proposed in our previous work which is based on cross terms, which were made zero in the conventional spectral subtraction method. This technique could effectively suppress musical noise, but at the cost of speech in-telligibility. To overcome this problem, a post pro-cessing technique is proposed in this paper, which is based on the gray level values of pixels in the spec-trogram of enhanced speech. This method can ef-fectively suppress musical noise without noticeable effect on speech intelligibility. Both subjective and objective performance assessments confirm that our method is effective