Sound Source Analysis for a Hearing Assistance Robot

摘要

To improve the quality of life of hearing-impaired people by using a hearing assistance robot, a sound source analysis system is proposed. Sound source localization and categorization are conducted and visualized for the users. Speech recognition is also conducted when the detected sound is judged as human voice. In this case, recognized speeches are shown on the screen as captions. Experimental results demonstrate the effectiveness of the proposed method.