Skip to Main Content
We propose an improved spectral subtraction method for reducing acoustic noise added to speech in colored noise environments like helicopter cockpit or car engine. This implementation uses over-subtraction method for spectral subtraction. The noise signal does not affect the speech signal uniformly over the whole spectrum. A nonlinear multi-band Bark scale frequency spacing approach was proposed to reduce colored noise. Simulations showed a better quality in terms of Itakura Saito distance and perceptual evaluation of quality for the enhanced speech.