Noise Reduction Algorithm Based On Signal-to-noise Ratio Of Electrical Signal For Cochlear Implants

Cochlear implant(CI)or electronic cochlea,is the only bionic instrument to restore the hearing of patients with severe hearing loss.It directly stimulates the auditory nerve by implanting an array of electrodes into patient’s cochlea to replace the missing hair cells in order to achieve auditory perception.CI is of great significance for patients with severe hearing loss in recovering their hearing and integrating into social life.At present,the technology of CI is relatively mature,and the existing CI system has been able to allow implanters to talk in quiet environment,but in noise environment,the speech intelligibility of implanters is still poor,which seriously restricts their social communication ability.Therefore,improving the speech perception ability of Cochlear implanters in noisy environment is one of the key issues in CI research.To deal with the abovementioned problem,a CI denoising algorithm based on signal-tonoise ratio(SNR)of electrical stimulation signal is proposed in this paper.Previous studies have shown that there are significant differences between the perception mechanism of electrical hearing and that of acoustic hearing.However,most of the existing denoising algorithms are based on the auditory research of normal hearing people or people with mild hearing loss.In this paper,by studying the mapping relationship between the SNR of acoustic signal and that of electrical stimulation signal of the CI system,the SNR of electrical stimulation signal is estimated.Based on this estimation,a denoising algorithm based on CI electrical hearing is proposed.In this paper,the performance of the proposed algorithm is evaluated by analyzing the signal envelope,spectrogram,the normalized covariance measure and the weighted spectral slope distance.Then,the speech reception threshold experiments and speech recognition experiments are conducted to verify the noise reduction performance of the proposed algorithm.The experimental results show that the proposed algorithm has better performance for CI system under noise conditions,but it is still poorer than that of noise reduction method based on acoustic SNR.Finally,based on these analysis results,this paper further discusses and verifies the possible causes of the analysis results.The further analysis shows that the estimation method of the SNR of the electrical stimulation signal has a significant effect on the performance of experiments.In general,the noise reduction algorithm proposed in this paper is of value.The work in this paper lays a foundation for the subsequent research on the signal-tonoise ratio denoising algorithm based on electrical stimulation signals.