Speech recognition in analog multichannel cochlear prostheses: initial experiments in controlling classifications

Abstract

Computer-synthesized vowels were used to examine methods for controlling and measuring the perceptions elicited during electrical stimulation of the human cochlea. In the first experiment, we measured the importance of the second formant (F2) in the identification of vowels, matched for duration, in a single subject with a multichannel cochlear implant. The subject never confused vowels having a "low" frequency F2 with those having a "high" frequency F2. In the second experiment, identification functions were generated for a series of vowels varying only in F2. When the pattern of F2 stimulation at the basilar membrane was manipulated, vowel identification functions were altered. For the categorization of vowels, the data indicate that the relative cochlear position of F2 stimulation was more important than fine-grain temporal waveform cues. The data are supportive of cochlear implant coding strategies that make use of cochlear place information. In the later experiments, we manipulated filter passbands and channel gains to explore their effect on these classifications. These preliminary studies indicate that it is possible to "fine-tune" such classifications.

Early UCSF contributions to the development of multiple-channel cochlear implants

“…After a significant period of effective use, in which as noted earlier, we documented speech reception abilities that substantially exceeded those achieved with single channel devices (Loeb, 1985(Loeb, , 1990White et al, 1990;Merzenich, 1985;Schindler et al, 1987;Schindler and Kessler, 1989), to our dismay, implantable connectors began to fail because of unanticipated dehission of ions across the narrow connector gaps under these high (>200atm) connector closure pressures.…”

“…In studies led by Robert Shannon, Elmer Owens and Marlene Ochs, we conducted parametric studies that provided an early foundation for understanding CI-generated sound and phonemic and speech perception (Shannon, 1983a;White et al, 1990;Merzenich, 1985Merzenich, , 1983bShannon, 1981;Shannon, 1985;. These studies documented interference and successivesignal masking effects that were highly informative for CI processor designs.…”

“…Among many other findings, we were the first group that showed, in animal then in human models, that we could "steer" stimulation to create "virtual" CI channels. For example, we could differentially bias adjacent channels and thereby systematically shift the boundaries between two confusable phonemes, or roll phonemic representations back and forth across two adjacent electrodes (White et al, 1990). Strategies to improve CI function are now appropriately strongly focused on generating more continuously coded representations by using more sophisticated forms of current steering, necessarily requiring bipolar (or tri-or quadripolar) stimulus control.…”

“…With our isolated-channel current-controlled analog stimulation models applied with "peak picking" stimulation (differentially engaging the 8th nerve array as a function of spectral energy) to reduce channel interference, we recorded higher speech reception abilities that had been recorded with single channel devices (White et al, 1990;Merzenich, 1985;Schindler et al, 1987;Schindler and Kessler, 1989). Still, we realized that devices would ultimately have to be implemented using digital pulsatile stimulation to further reduce channel interference.…”

Effects of Channel-to-Electrode Mappings on Speech Reception with the Ineraid Cochlear Implant

“…White, Ochs, Merzenich, and Schubert (1990) tested the effects of reversing the two intermediate channels of a custom four-channel analog processor (similar to the standard Ineraid processor) in a single subject implanted with the UCSF electrode (Loeb et al, 1983). The stimuli were 10 steadystate (250 msec) synthetic vowels in which the second formant varied evenly from 1150 to 2050 Hz and all other aspects of the signals remained con-EAR & HEARING / OCTOBER 1995 stant, thereby forming an /u/-/i/ continuum.…”