Integration of Anomaly Machine Sound Detection into Active Noise Control to Shape the Residual Sound

An active noise control (ANC) system generates a secondary sound to destructively interfere with the undesirable noise. Existing ANC algorithms are mainly designed to minimize the power of the residual sound, with few considerations to the listening experience. This results in a pressing issue in practice whereby the residual sound is perceived to be different from the undesirable noise. When the ANC system is deployed to reduce the noise level in a factory environment, workers may feel strange because they are used to detecting the anomaly machine sound by their auditory perception. In order to solve this problem, this paper proposes to integrate anomaly sound detection (ASD) into the ANC system in order for the residual sound to represent the same machine status as the original machine noise. The ASD module is used to simulate human judgement. A homothety constrained ANC algorithm is developed to synchronously reduce the sample-wise power and keep the segment-wise machine status of the residual sound. The experiment results validate the effectiveness of the homothety constrained ANC algorithm in noise reduction, and the subjective test results show that the ASD-integrated ANC system results in less confusing perceptions of the residual sound.