Cochlear Implants: Consequences of Microphone Aging on Speech Recognition
Aging is a general phenomenon which affects everything and everybody in this world. Designed for the rehabilitation of profound deafness, cochlear implants (CI) do not escape to this general rule. One very insidious effect concerns the microphone as an ongoing drift occurs over the time. In this work we wish to assess the consequences of this evolution on speech recognition. In order to perform this task, a general population of CI users and NH subjects (using a CI simulator) participated to this study. They listened to French dissyllabic words and we recorded recognition percentages. Words were presented to the listeners in noise with a variable signal to noise ratio (SNR) and the percentages ranged from 0 to 100%. For the CI simulator, the drift was simulated from data coming from figures measured on regular hearing aids. This choice seems relevant as CIs and hearing aids use the same microphones. Also, the CI simulator we used, picked up the general principles of a vocoder to represent the classical coding strategies used in CIs (CIS-like and n-of-m). With CI users, the results were compared before and after cleaning the microphones; also, in a subgroup of CI users, we performed the replacement of the head filter protecting the microphone and the recognition percentages were compared with those coming from the standard “Brush and Blow” cleaning procedure. The results have been revisited and quantified after a curve fitting. The outcomes indicated that the CIS-like coding schemes were less sensitive to aging than the n-of-m strategies. Also, cleaning ameliorated the recognition performances, but the increase was not dramatically high. Furthermore, the improvement mainly occurred in the middle of the SNR range where the noise was not too intense. We made these observations with CI users and with NH subjects so it indicates that the results should be linked to the properties of the signal. Finally, as we cannot stop the consequences of aging, we can set up an action plan to reduce its effect. And this is true in everyday life. In the case of CIs, a lot of solutions are available, among them the choice of the sound coding strategy and the periodicity of the clinical check and device setting.
KeywordsCochlear implants Microphone aging CI and NH listeners Coding strategies Cleaning procedures Syllable recognition in noise
The authors are grateful to the persons and the institutions who participated to the study; M. Kevin Perreaut who initiated the work, Dr. Fabien Seldran and Dr. Fabien Millioz for the scientific contribution and Ms. Evelyne Veuillet for the links with the ethic committee. We also wanted to thank the members of the CRIC Lyon and the staff of the Edouard-Herriot hospital for their collaboration, the subjects who listened to the Fourier’s lists, the Hospitals of Lyon and the Polytechnic School of the University of Lyon.
- 1.Aguiar, D.E., Taylor, N.E., Li, J., Gazanfari, D.K., Talavage, T.M., Laflen, J.B., et al.: Information theoretic evaluation of a noiseband-based cochlear implant simulator. Hear. Res. (2015)Google Scholar
- 7.Razza, S., Burdo, S., Bonaretti, S.: Acoustical signal check: microphone integrity evaluation through a common hearing aid analyzer. In: 5th Object Measurement Symposium Cochlear Implants ABI, pp. 19–22 (2007)Google Scholar
- 10.Wilson, B.S., Lawson, D.T., Zerbi, M., Finley, C.C., Wolford, R.D.: New processing strategies in cochlear implantation. Am. J. Otol. 16, 669–675 (1995)Google Scholar
- 12.Dorman, M.F., Loizou, P.C., Fitzke, J., Tu, Z.: Recognition of monosyllabic words by cochlear implant patients and by normal-hearing subjects listening to words processed through cochlear implant signal processing strategies. Ann. Otol. Rhinol. Laryngol. Suppl. 185, 64–66 (2000)CrossRefGoogle Scholar
- 13.Shannon, R.V., Fu, Q.-J., Galvin, J.: The number of spectral channels required for speech recognition depends on the difficulty of the listening situation. Acta Oto-Laryngol. Suppl. 50–54 (2004)Google Scholar
- 14.Dorman, M.F., Loizou, P.C.: The identification of consonants and vowels by cochlear implant patients using a 6-channel continuous interleaved sampling processor and by normal-hearing subjects using simulations of processors with two to nine channels. Ear Hear. 19, 162–166 (1998)CrossRefGoogle Scholar
- 16.Serra, P.-O.: Effet de l’entretien des aides auditives sur leurs performances [Audiology diploma dissertation]. University of Montpellier (2015)Google Scholar
- 17.Perreaut, K., Gallego, S., Berger-Vachon, C., Millioz, F.: Influence of microphone encrusting on the efficiency of cochlear implants preliminary study with a simulation of CIS and “n-of-m” strategies. AMSE J. Model. C 75–2, 199–208 (2014)Google Scholar
- 24.Jeanvoine, A., Gnansia, D., Truy, E., Berger-Vachon, C.: Contribution of noise reduction algorithms: perception versus localization simulation in the case of binaural cochlear implant (BCI) coding. Emerg. Trends Comput. Biol. Bioinf. Syst. Biol. 307–324 (2015)Google Scholar