Optimization of a Spectral Contrast Enhancement Algorithm for Cochlear Implants Based on a Vowel Identification Model.

Speech intelligibility achieved with cochlear implants (CIs) shows large variability across different users. One reason that can explain this variability is the CI user's individual electrode nerve interface which can impact the spectral resolution they can achieve. Spectral resolution has been reported to be related to vowel and consonant recognition in CI listeners. One measure of spectral resolution is the spectral modulation threshold (SMT), which is defined as the smallest detectable spectral contrast in a stimulus. In this study we hypothesize that an algorithm that improves SMT may improve vowel identification, and consequently produce an improvement in speech understanding for CIs. With this purpose we implemented an algorithm, termed spectral contrast enhancement (SCE) that emphasizes peaks with respect to valleys in the audio spectrum. This algorithm can be configured with a single parameter: the amount of spectral contrast enhancement entitled "SCE factor". We would like to investigate whether the "SCE factor" can be individualized to each CI user. With this purpose we used a vowel identification model to predict the performance produced by the SCE algorithm with different "SCE factors" in a vowel identification task.In five CI users the new algorithm has been evaluated using a SMT task and a vowel identification task. The tasks were performed for SCE factors of 0 (no enhancement), 2 and 4. In general it seems that increasing the SCE factor produces a decrease in performance in both the SMT threshold and vowel identification.

Spectral contrast enhancement improves speech intelligibility in noise for cochlear implants.

Spectral smearing causes, at least partially, that cochlear implant (CI) users require a higher signal-to-noise ratio to obtain the same speech intelligibility as normal hearing listeners. A spectral contrast enhancement (SCE) algorithm has been designed and evaluated as an additional feature for a standard CI strategy. The algorithm keeps the most prominent peaks within a speech signal constant while attenuating valleys in the spectrum. The goal is to partly compensate for the spectral smearing produced by the limited number of stimulation electrodes and the overlap of electrical fields produced in CIs. Twelve CI users were tested for their speech reception threshold (SRT) using the standard CI coding strategy with and without SCE. No significant differences in SRT were observed between conditions. However, an analysis of the electrical stimulation patterns shows a reduction in stimulation current when using SCE. In a second evaluation, 12 CI users were tested in a similar configuration of the SCE strategy with the stimulation being balanced between the SCE and the non-SCE variants such that the loudness perception delivered by the strategies was the same. Results show a significant improvement in SRT of 0.57 dB (p < 0.0005) for the SCE algorithm.

Remixing music using source separation algorithms to improve the musical experience of cochlear implant users.

Music perception remains rather poor for many Cochlear Implant (CI) users due to the users' deficient pitch perception. However, comprehensible vocals and simple music structures are well perceived by many CI users. In previous studies researchers re-mixed songs to make music more enjoyable for them, favoring the preferred music elements (vocals or beat) attenuating the others. However, mixing music requires the individually recorded tracks (multitracks) which are usually not accessible. To overcome this limitation, Source Separation (SS) techniques are proposed to estimate the multitracks. These estimated multitracks are further re-mixed to create more pleasant music for CI users. However, SS may introduce undesirable audible distortions and artifacts. Experiments conducted with CI users (N = 9) and normal hearing listeners (N = 9) show that CI users can have different mixing preferences than normal hearing listeners. Moreover, it is shown that CI users' mixing preferences are user dependent. It is also shown that SS methods can be successfully used to create preferred re-mixes although distortions and artifacts are present. Finally, CI users' preferences are used to propose a benchmark that defines the maximum acceptable levels of SS distortion and artifacts for two different mixes proposed by CI users.

Positron Emission Tomography Imaging Reveals Auditory and Frontal Cortical Regions Involved with Speech Perception and Loudness Adaptation.

Considerable progress has been made in the treatment of hearing loss with auditory implants. However, there are still many implanted patients that experience hearing deficiencies, such as limited speech understanding or vanishing perception with continuous stimulation (i.e., abnormal loudness adaptation). The present study aims to identify specific patterns of cerebral cortex activity involved with such deficiencies. We performed O-15-water positron emission tomography (PET) in patients implanted with electrodes within the cochlea, brainstem, or midbrain to investigate the pattern of cortical activation in response to speech or continuous multi-tone stimuli directly inputted into the implant processor that then delivered electrical patterns through those electrodes. Statistical parametric mapping was performed on a single subject basis. Better speech understanding was correlated with a larger extent of bilateral auditory cortex activation. In contrast to speech, the continuous multi-tone stimulus elicited mainly unilateral auditory cortical activity in which greater loudness adaptation corresponded to weaker activation and even deactivation. Interestingly, greater loudness adaptation was correlated with stronger activity within the ventral prefrontal cortex, which could be up-regulated to suppress the irrelevant or aberrant signals into the auditory cortex. The ability to detect these specific cortical patterns and differences across patients and stimuli demonstrates the potential for using PET to diagnose auditory function or dysfunction in implant patients, which in turn could guide the development of appropriate stimulation strategies for improving hearing rehabilitation. Beyond hearing restoration, our study also reveals a potential role of the frontal cortex in suppressing irrelevant or aberrant activity within the auditory cortex, and thus may be relevant for understanding and treating tinnitus.

Improving neural decoding in the central auditory system using bio-inspired spectro-temporal representations and a generalized bilinear model.

We study the impact of different encoding models and spectro-temporal representations on the accuracy of Bayesian decoding of neural activity recorded from the central auditory system. Two encoding models, a generalized linear model (GLM) and a generalized bilinear model (GBM), are compared along with three different spectro-temporal representations of the input stimuli: a spectrogram and two bio-inspired representations, i.e. a gammatone filter bank (GFB) and a spikegram. Signal to noise ratios between the reconstructed and original representations are used to evaluate the decoding, or reconstruction accuracy. We experimentally show that the reconstruction accuracy is best with the spikegram representation and worst with the spectrogram representation and, furthermore, that using a GBM instead of a GLM significantly increases the reconstruction accuracy. In fact, our results show that the spikegram reconstruction accuracy with a GBM fitting yields an SNR that is 3.3 dB better than when using the standard decoding approach of reconstructing a spectrogram with GLM fitting.

Investigation of a new electrode array technology for a central auditory prosthesis.

Ongoing clinical studies on patients recently implanted with the auditory midbrain implant (AMI) into the inferior colliculus (IC) for hearing restoration have shown that these patients do not achieve performance levels comparable to cochlear implant patients. The AMI consists of a single-shank array (20 electrodes) for stimulation along the tonotopic axis of the IC. Recent findings suggest that one major limitation in AMI performance is the inability to sufficiently activate neurons across the three-dimensional (3-D) IC. Unfortunately, there are no currently available 3-D array technologies that can be used for clinical applications. More recently, there has been a new initiative by the European Commission to fund and develop 3-D chronic electrode arrays for science and clinical applications through the NeuroProbes project that can overcome the bulkiness and limited 3-D configurations of currently available array technologies. As part of the NeuroProbes initiative, we investigated whether their new array technology could be potentially used for future AMI patients. Since the NeuroProbes technology had not yet been tested for electrical stimulation in an in vivo animal preparation, we performed experiments in ketamine-anesthetized guinea pigs in which we inserted and stimulated a NeuroProbes array within the IC and recorded the corresponding neural activation within the auditory cortex. We used 2-D arrays for this initial feasibility study since they were already available and were sufficient to access the IC and also demonstrate effective activation of the central auditory system. Based on these encouraging results and the ability to develop customized 3-D arrays with the NeuroProbes technology, we can further investigate different stimulation patterns across the ICC to improve AMI performance.

Neural representation in the auditory midbrain of the envelope of vocalizations based on a peripheral ear model.

The auditory midbrain implant (AMI) consists of a single shank array (20 sites) for stimulation along the tonotopic axis of the central nucleus of the inferior colliculus (ICC) and has been safely implanted in deaf patients who cannot benefit from a cochlear implant (CI). The AMI improves lip-reading abilities and environmental awareness in the implanted patients. However, the AMI cannot achieve the high levels of speech perception possible with the CI. It appears the AMI can transmit sufficient spectral cues but with limited temporal cues required for speech understanding. Currently, the AMI uses a CI-based strategy, which was originally designed to stimulate each frequency region along the cochlea with amplitude-modulated pulse trains matching the envelope of the bandpass-filtered sound components. However, it is unclear if this type of stimulation with only a single site within each frequency lamina of the ICC can elicit sufficient temporal cues for speech perception. At least speech understanding in quiet is still possible with envelope cues as low as 50 Hz. Therefore, we investigated how ICC neurons follow the bandpass-filtered envelope structure of natural stimuli in ketamine-anesthetized guinea pigs. We identified a subset of ICC neurons that could closely follow the envelope structure (up to ~100 Hz) of a diverse set of species-specific calls, which was revealed by using a peripheral ear model to estimate the true bandpass-filtered envelopes observed by the brain. Although previous studies have suggested a complex neural transformation from the auditory nerve to the ICC, our data suggest that the brain maintains a robust temporal code in a subset of ICC neurons matching the envelope structure of natural stimuli. Clinically, these findings suggest that a CI-based strategy may still be effective for the AMI if the appropriate neurons are entrained to the envelope of the acoustic stimulus and can transmit sufficient temporal cues to higher centers.