Graeme Clark Collection
Permanent URI for this collection
Search Results
1 - 4 of 4
-
ItemSignal processing for multichannel cochlear implants: past, present and future [Abstract]( 1994)Since the late 1970's, many groups have worked on developing effective signal processing for multichannel cochlear implants. The main aim of such schemes has been to provide the best possible speech perception for those using the device. Secondary aims of providing awareness and discrimination of environmental sounds and appreciation of music have also been considered. Early designs included some that attempted to simulate the normal cochlea. The application of such complex processing schemes was limited by the technology of the times. In some cases, researchers reverted to the use of single channel systems which could be controlled reliably with the existing technology. In other cases, as with the Australian implant, a simple multichannel processing scheme was devised that allowed a reliable implementation with available electronics. Over the next 15 years, largely due to the improvements in integrated circuit technology, the signal processors have slowly become more complex. Further psychophysical research has shown how additional information can be transferred effectively to implant users via electrical stimulation of the cochlea. This has lead to rapid improvement in the speech perception abilities of adults using cochlear implants. Some of the main developments in signal processing over the last 15 years will be discussed along with the latest speech perception results obtained with the new SPEAK processing scheme for the Australian 22-channel cochlear implant. Initial results for SPEAK show mean scores of 70% (equivalent to 85-90% phoneme scores) for open set monosyllabic word testing for experienced adult users. Although there remains a large range of performance for all users of cochlear implants, average speech perception scores for all implanted adults have also improved significantly with the developments in signal processing. It appears likely that multichannel cochlear implants will be a viable alternative for the treatment of severe hearing loss in the future.
-
ItemComparison of current speech coding strategies( 1993)This paper reports on two studies carried out at the University of Melbourne jointly with Cochlear Pty Ltd. The studies demonstrated substantial speech perception improvements over the current Multipeak strategy in background noise.
-
ItemMultichannel cochlear implantation in children: a summary of current work at The University of Melbourne( 1991)This paper summarizes research work relating to multichannel cochlear implantation in children at the University of Melbourne. Ongoing safety studies relating to the implantation of young children are discussed. Results of these studies suggest that special design considerations are necessary for a prosthesis to be implanted in children under the age of 2 years. Results of clinical assessment of implanted children and adolescents are also discussed in terms of speech perception, speech production, and language development, and some possible predictive factors are suggested. Preliminary data suggests that a high proportion of young children can achieve open-set speech perception with the cochlear implant given appropriate training and support. Initial results with adults using new speech processing hardware and a new coding scheme are also presented. These suggest that improved speech perception in quiet and competing noise is possible with the new system.
-
ItemInitial speech perception results with the new multipeak speech processor for the 22-electrode cochlear prosthesis [Abstract]( [1990])A new speech processor has been developed for the 22-electrode cochlear prosthesis by Cochlear Pty Ltd working in conjunction with the Department of Otolaryngology at the University of Melbourne. The new device, known as the MSP, combines smaller, more efficient hardware with a new speech coding scheme in an attempt to provide better speech perception in everyday environments for implant users. The MSP operates with the current implant device so there is no need for existing implantees to have revision surgery to make use of the new development. The multipeak speech coding scheme, which has been implemented in the MSP, provides information from three high frequency spectral bands, in addition to the parameters of voice pitch, amplitude and first and second formants which have been provided in the existing FOFIF2 coding scheme for the last four years. Initial speech perception results with research subjects have shown significant improvements in performance for the MSP over the older system (WSP III). The most encouraging result is that open-set speech perception in the presence of competing noise has improved substantially. For example, mean scores for BKB sentences in a 10 dB signal-to-noise ratio were 64% for the MSP and 31% for the WSP III. Further investigations have shown that both the hardware improvements and the new multipeak speech coding scheme have contributed significantly to the overall improvement in performance. Studies are continuing to analyse further the potential of the new system.