Skip to Main Content
In this paper, a new smart noise reduction system for sinusoidal and wideband noise is proposed. We have proposed the cascaded structure of adaptive line enhancer (ALE) and noise reconstruction system (NRS). However, it has the problem that the quality of the enhanced speech signal is degraded. The proposed system uses two ALEs and a noise reconstruction filter (NRF). First, ALE1 and ALE2 are used to estimate the sinusoidal noise and the speech signal, respectively. However, the ALE cannot estimate the unvoiced speech, so the quality of the speech estimated by the ALE is degraded. Therefore, the NRF is used to improve the quality of the enhanced speech. The NRF accurately estimates the background noise from the signal obtained by subtracting estimated speech signal from the noisy speech signal. Noise reduction is achieved by subtracting the reconstructed noise from the noisy speech signal. Simulation results show the effectiveness of the proposed smart noise reduction system.
Circuits and Systems, 2004. MWSCAS '04. The 2004 47th Midwest Symposium on (Volume:2 )
Date of Conference: 25-28 July 2004