Shaikat Hossain - Improving sound coding for cochlear implant users

Cochlear implants (CI) are amazing if all you care about is speech. They convert audio into electrical impulses which are fed into a user's cochlea. Young users can learn spoken language using only a CI. But cochlear implants are really lousy in complicated audio environments, and at representing music. One of the problems is that they large encode the overall speech spectrum, but ignore the details (like timing) that are important for understanding the world around us. What can we do better?

Can we provide better electrical stimuli so CI users can enjoy music and they can understand speech in noise?

Who: Shaikat Hossain
What: Improving sound coding for cochlear implant users
When: Friday February 17th at 10:30AM
Where: CCRMA Seminar Room
Why: We'd love to know how to make cochlear implants work much better.  Perhaps temporal coding is the key.

Title: Improving sound coding for cochlear implant users

Abstract: This presentation will provide an overview of how cochlear implant (CI) devices work and explore some of the limitations of contemporary CI sound processing strategies. Findings from psychophysical studies will be presented which focus on ways of improving the encoding of temporal fine structure information. Implications for the development of novel sound coding strategies which could lead to improved speech in noise and music perception abilities for CI users will be explored.

Biography: Shaikat Hossain works as an acoustic engineer in the areas of spatial audio and music technology. Prior to joining Apple, he was a researcher in the area of cochlear implants focused on improving spatial hearing and music perception abilities of CI users.  See this paper of his: Perceptual learning of pitch provided by cochlear implant stimulation rate