PRESERVING SPECTRAL CONTRAST IN AMPLITUDE COMPRESSION FOR HEARING AIDS
Janet Rutledge
UMBC
Digital hearing aids provide the opportunity for introducing new customized processing strategies for individual listeners with hearing loss. Recent attempts to customize digital signal processing algorithms have not been fully satisfactory in speech quality nor proven to improve speech understanding. For example, multichannel amplitude compression processing is used to reduce the level variations of speech to fit the reduced dynamic ranges of listeners with sensorineural hearing loss. This processing, however, can result in smearing of temporal information, artifacts due to spectral discontinuities at fixed channel edges, and spectral flattening due to reduced peak-to-valley ratios. Waveform parameterization models, such as wavelets or sinusoidal models, can be used in place of filter-based techniques. These models allow greater flexibility in the range of digital signal processing, and lend themselves to time-varying techniques.
We present an amplitude compression algorithm based on a sinusoidal speech model that preserves the important spectral peaks. The algorithm operates on a time-varying, stimulus-dependent basis to adjust to the speech variations and the listener's hearing profile. The algorithm provides fast-acting compression without artifacts, has time-varying frequency channels, and is computationally inexpensive. Preliminary subject tests indicate benefit from real-time Col-SM processing that is greater than that from fixed-frequency multichannel compression without some of the undesirable artifacts. This method could be extended to provide real-time enhancement of spectral contrast.
Dr. Janet C. Rutledge received the B.S. degree in electrical engineering from Rensselaer Polytechnic Institute in 1983, and the M.S. and Ph.D. in electrical engineering from the Georgia Institute of Technology in 1984 and 1990, respectively. She is currently the Associate Dean of the Graduate School at the University of Maryland, Baltimore County (UMBC), and has affiliate faculty appointments in the Computer Science and Electrical Engineering Department at UMBC and in the School of Medicine at the University of Maryland, Baltimore (UMAB). Dr. Rutledge came to UMBC from the National Science Foundation where she served as the Program Director for the Graduate Research Fellowship Program, and Program Director in the Division of Engineering Education and Centers, and in the Division of Undergraduate Education. She held a faculty position in the Electrical and Computer Engineering Department at Northwestern University. Dr. Rutledge's primary research area is modeling and compensating for the effects of sensorineural hearing loss and other communication disorders. She is the author of numerous journal and conference publications, an undergraduate textbook, and holds a patent. She has held several leadership positions in the IEEE, and is a member of the Georgia Tech Engineering Advisory Board and the Rensselaer Polytechnic Institute Board of Trustees.
Host: Joel Morris
Department of Computer Science & Electrical Engineering
All Rights Reserved. Comments/Suggestions to the Site Editor.