PubMed HealthSearch

PubMed · 2356719

Wearable digital speech processor for cochlear implants using a TMS320C25.

Abstract

Based on a single-chip DSP (TMS320C25, Texas Instruments) a programmable battery-operated sound processor with a digital encoder interface for the Nucleus-22 cochlear implant (CI) was built. The number of quasi-simultaneously addressed electrodes is only limited by the selected pulse width and the maximum rate of stimulation and can be as high as 10 electrodes at 300 Hz repetition rate. Implementation of various processing strategies (formant or channel vocoder, filterbank, zero crossings, etc.) is possible as well as sophisticated adaptive noise reduction. Programs and stimulation parameters are stored in electrically erasable memory and may be updated via a host computer. The built-in analog output may be used for single-channel stimulation or acoustic verifications of the sound processing algorithms. The power consumption with current 16K word data memory and at maximum stimulation rate is about 1 Watt, which necessitates recharging of batteries after 11 h.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

N Dillier, C Senn, T Schlatter, M Stöckli, U Utzinger. 1990. Wearable digital speech processor for cochlear implants using a TMS320C25.. https://pubmed.ncbi.nlm.nih.gov/2356719/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

[Cochlear implantation].

Since the Israeli cochlear implant program was started in June 1989, 74 candidates have been evaluated for implantation. The main causes for hearing loss were meningitis (16 patients), aminoglycoside toxicity (9), congenital (6), viral (e.g. mumps) (6), and middle ear infection. Candidates were rejected mainly due to psychosocial factors (40%) and audiological parameters (30%). 16 postlingual deaf were implanted. In 7 of them (43.7%) the results were excellent, in 4 (25.0%) good, in 3 (18.7%) fair; 1 case is currently in a training program and 1 is a failure. There were 2 major complications: facial nerve palsy which necessitated surgical decompression, and insertion of the electrodes into the hypotympanic air cells, corrected by reoperation.

Cochlear Implants

Performance over time with a nucleus or Ineraid cochlear implant.

This investigation determined whether the audiological performance of cochlear implant users varied with experience. Thirteen Nucleus and 14 Ineraid subjects were evaluated at 1, 9, and 18 mo after cochlear implant connection. Ten Nucleus and five Ineraid subjects were tested at 30 mo. On average, the ability of the subjects to recognize words and phonemes in an audition-only condition improved during the first 9 mo, as did their ability to recognize spondees in noise. The phoneme scores continued to improve during the next 9 mo. Environmental sound recognition improved gradually; significant improvement from the 1 mo scores was not noted until 18 mo. About half of the subjects who demonstrated poor word recognition at 1 mo showed significantly improved percent word correct scores by 18 mo. The Nucleus and Ineraid subjects did not differ in their patterns of change over time. An information transmission analysis performed on the subjects' consonant confusion matrices showed relatively little change for the nasality and place features during the first 18 mo, and relatively large change for the voice, duration, and frication features. Most improvement in the feature scores occurred during the first 9 mo.

Cochlear Implants