This application proposes to investigate, evaluate and refine a promising novel digital technique for reducing wind noise in hearing aids. The long-term objective is to develop the invention into the method of choice for wind noise reduction in commercial digital hearing aids. Wind noise is highly disturbing to hearing impaired individuals who wish to participate in outdoor conversations or conversations during activities such as walking or running. In these situations, wind noise significantly reduces signal-to-noise ratio and, consequently, the intelligibility of speech and sounds is significantly impaired. Based upon work started at Bell Labs and continued at MH Acoustics, we have recently conceived and developed a novel and efficient method for reducing wind noise. The basic theory has been developed and the algorithm has been implemented and verified in off-line batch processing. The next steps in the development process are to: a) verify and refine our method for detecting and suppressing wind noise. b) Build a controlled acoustic environment for testing of the method's capability. c) Modify a commercial hearing aid to interface to an external measurement/processing system. d) Implement our algorithm in real-time on a commercial processing platform e) refine the algorithm's performance under real-world conditions.
