NEURAL SPEECH ENHANCEMENT

Demos | Research | Database and Tools | Publications | People | Links
Speech enhancement is concerned with the processing of corrupted or noisy speech in order to improve the quality or intelligibility of the signal. Applications range from front-ends for speech recognition systems, to enhancement of telecommunications for aviation, military, teleconferencing, and cellular environments. At OGI our goal is to develop state-of-the art systems and conduct fundamental research based on nonlinear signal processing, machine learning, and perceptual techniques.
General research on nonlinear signal processing and neural networks
Review chapter on-line: Networks For Speech Enhancement (ps version)
Principle Investigator: Eric Wan
Graduate Students: Houwu Bai, Rudolph van der Merwe , Alex T. Nelson (graduated 2000)