Generation of Acoustic Stimuli from a Modeled Cochlear Implant-Induced Neural Activation Pattern
- Huang, Laflen, and Talavage
Cochlear implants are electronic devices that have enabled individuals with severe to profound hearing losses to regain some hearing. Nearly all of those who receive cochlear implants (CIs) regain the sensation of sound. There is, however, substantial variability in speech perception performance among users of cochlear implants (Staller et al., 1997). One factor that may contribute to the individual differences in performance is the speech processing strategy used. Traditionally, researchers implemented experimental real time strategies on specialized digital signal processors (DSPs). This approach requires specialized programming knowledge and consequently can be expensive. In this respect, we have simplified the development of real time processing strategies by using an IBM compatible laptop computer to perform all signal processing, and have implemented only a few interface functions on a DSP. Both the continuous interleaved sampling (CIS) and n-of-m speech processing schemes were implemented on a PC in C++. In addition to encoding and transferring speech cues to a cochlear implant, the processor described can also implement an acoustic noiseband based model of a cochlear implant for presentation to listeners with normal hearing.