High Definition Digital Signal Processing

A speech processing strategy, or speech coding strategy, is a code that converts sound waves into patterns of electrical pulses. These electrical pulses are generated by the implant to stimulate the auditory nerve via an electrode array within the cochlea. The TEMPO+ Speech Processor uses Continuous Interleaved Sampling and the Hilbert Transform to provide High Definition Digital Signal Processing, referred to as CIS+.

Continuous Interleaved Sampling (CIS) is a pulsatile (digital) speech processing strategy. Among various cochlear implants, there are a number of CIS coding strategies. While these strategies share the CIS name, each cochlear implant system implements the CIS strategy differently. According to the original developers of CIS, accurate implementation of the strategy fulfills the following criteria(1):

:: uses digital filters and fast sampling
:: can stimulate at high rates
:: provides new information in each pulse
:: implements technical parameters correctly
:: has flexible stimulation parameters
:: ensures that stimulation pulses are non-overlapping

CIS+, available only in the TEMPO+ Speech Processor, adheres to strict guidelines for optimal implementation of the CIS strategy. Additionally, CIS+ provides a wider frequency range and highly flexible stimulation parameters. The TEMPO+ microphone is capable of a frequency response from 200 to 10,000 Hz. Fitting parameters that can be customized to
meet the individual needs of the cochlear implant user include:

:: threshold and most comfortable levels
:: pulse duration
:: frequency range
:: frequency bands
:: bandpass assignment
:: number of channels
:: stimulation rate
:: stimulation order
:: maplaw/compression function
:: volume mode
:: volume range

CIS+ and the TEMPO+ also use a highly advanced mathematical algorithm, the Hilbert Transform, to provide High Definition Digital Signal Processing. The Hilbert Transform is a very precise equation, which results in a stimulation pattern that very closely matches the overall envelope (acoustic shape) of the original sound. Other cochlear implant systems use techniques called “wave rectification and low-pass filtering” or “fast Fourier transform” to represent sound, but the Hilbert Transform tracks incoming sounds much more accurately for the most mathematically detailed representation of the original sound. A comparative study with various cochlear implant systems revealed that the MED-EL TEMPO+ provides excellent performance on measures of speech perception in quiet and in noise.(2)

1 Wilson BS, Finley CC, Lawson DT, Wolford RD, Eddington DK, Rabinowitz WM. (1991). Better speech recognition with cochlear implants, Nature, Jul 18:352(6332);236-8.
2 Spahr AJ, Dorman MF. (2003). A comparison of performance among patients fit with the CII Hi-Resolution, 3G and TEMPO+ processors. Poster presentation at the Conference on Implantable Auditory Prostheses, August 17-22, Pacific Grove, CA.