Noise Reduction and Increased VAD Accuracy Using Spectral Subtraction
Jeffery J. Faneuff, Donald R. Brown · 2003
This paper shows that performing voice activity detection (VAD) on the output of a spectral subtraction noised reduced signal increases the accuracy of the VAD and reduces the VAD sensitivity to fixed thresholds. An initial VAD decision is used to control the noise estimate update in the spectral subtraction algorithm. The more accurate VAD after the first spectral subtraction is then used to reprocess the original noisy speech again via spectral subtraction to reduce the noise while not attenuating the speech. Auditory masking thresholds were used to weight the spectral subtraction to avoid the introduction of musical noise artifacts.