Abstract
In this paper, we propose a novel blind source separation method for the hose-shaped rescue robot based on independent low-rank matrix analysis and statistical speech enhancement. The rescue robot is aimed to detect victims'speech in a disaster area, wearing multiple microphones around the body. Different from the common microphone array, the positions of microphones are unknown, and the conventional beamformer cannot be utilized. In addition, the vibration noise (ego-noise) is generated when the robot moves, yielding the serious contamination in the observed signals. Therefore, it is important to eliminate the ego-noise in this system. This paper describes our newly developed software and hardware system of blind source separation for the robot noise reduction. Also, we report objective and subjective evaluation results showing that the proposed system outperforms the conventional methods in the source separation accuracy and perceptual sound quality via experiments with actual sounds observed in the rescue robot.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call