Machine learning to enhance temporal coding for cochlear implants

TEMPORAL: Machine learning to enhance temporal coding for cochlear implants

TEMPORAL combines academic research and medical expertise with the industry to achieve better communication and social participation for deaf and hard of hearing people. It is a collaboration of the (also internationally renowned) referral care (‘top referente zorg’, TRF) centre for cochlear implants (CI) at LUMC, the Leiden Institute of Advanced Computer Science, and Advanced Bionics, one of the world’s leading manufacturers of cochlear implants.

According to the World Health Organization (WHO), more than 5% of the world's population suffers from hearing loss. With increasing life expectancy, the number of severely hearing impaired will also rise. Today, electrical inner ear prostheses (cochlear implants, CIs) are the standard treatment for deaf and severely hard of hearing people. This allows adults to regain many of their communication capabilities, giving them better opportunities to remain socially active. In children, the CI enables the development of spoken language. This gives them better opportunities for education, work and social integration. However, the CI has a significant limitation in properly conveying detailed time information of sound to the auditory nerve. This manifests itself in difficulties with speech understanding in (fluctuating) background noise, tonal languages, directional hearing and music perception.

TEMPORAL will develop improved speech coding strategies for the CI speech processor that better transmit time information to the auditory nerve. A combination of three techniques is used for this: computer modeling, artificial intelligence (AI) and patient-related research. A distinction is made between hearing with two CIs (bilateral CIs) and with a CI and a hearing aid on the other ear (bimodal hearing).

These new speech coding strategies will lead to better perception in the aforementioned difficult listening situations. This will lead to better communication and social participation for deaf and hard of hearing people, resulting in a better quality of life.

Logo

Summary
TEMPORAL will develop improved speech coding strategies for the cochlear implant speech processor; these strategies will focus on improving the temporal features of sound for cochlear implant users. A combination of three techniques will be used: computer modeling, artificial intelligence and patient related research.
Technology Readiness Level (TRL)
2 - 7
Time period
48 months
Partners
Logo
Logo
Logo