A formant-estimating speech processor for cochlear implant patients

Abstract

A simple formant-estimating speech processor has been developed to make use of the “hearing” produced by electrical stimulation of the auditory nerve with a multiple-channel cochlear implant. Thirteen implant patients were trained and evaluated with a processor that presented the second formant frequency, fundamental frequency, and amplitude envelope of the speech (F0F2). Nine patients were trained and evaluated with a processor that presented the first and second formant frequencies, fundamental frequency, and first and second formant amplitudes (F0F1F2). The most common use of the speech processor was in conjunction with lipreading, so the patients were trained in lipreading plus hearing, as well as hearing alone. The F0F1F2 group performed significantly better in discrimination tasks and word and sentence recognition through hearing alone. The F0F1F2 group also showed a significantly greater improvement when hearing and lipreading was compared with lipreading alone in a speech tracking task. A study of spondee recognition in noise with hearing alone indicated that the added first formant information produced an improvement that was equivalent to a 5 dB increase in the signal-to-noise ratio.