Modular Implementation Of The Mobile End Of The Cochlear Speech Processing Function

Cochlear implant(CI),also known as electronic cochlea,is one of the most successful neuro prosthetic systems for human.It replaces acoustic stimulation by electrical stimulation to help patients with severe hearing problems restore partial hearing ability and this is the only effective way to treat patients with deafness.A CI device contains two parts,i.e.,the implant parts including a signal receiver,a stimulator,an electrode array and the outer parts including a microphone,a signal processor,a signal transmitter.The outer parts plays the role of signal acquisition,processing and transmission,especially in which the speech processing module is a key factor affecting the speech recognition ability of CI implanted patients.Limited by its computational performance,power consumption,volume and other factors,the existing outer parts can not provide enough computing resources required by efficient but complex speech processing algorithms.Thereby,it can not help implanted patients improve the speech recognition ability in complex noise environments.Moreover,the need of changing battery frequently and the large volume of the implant also brought great inconvenience,and even psychological pressure for users.This paper presents an improved scheme for the above-mentioned shortcomings of CI devices: cancel the signal acquisition and speech processing module of the outer parts,and this part of the function will be transplanted to the mobile terminal to achieve;The outer parts only contains the signal transmission module,including a new bluetooth module and the original radio frequency module,respectively,for signal transmission between the outer parts and mobile terminals and the implant parts.The ideas in this paper are realized on Android system: pick up speech signals by microphone and use computing resources of the mobile phone to process signals(CIS which is the most widely used speech coding strategy is adopted in this paper).The output of the electrical signal will be sent to the bluetooth module on the outer parts,and then sent to the implant parts via transcutaneous link in the form of electromagnetic waves.The high computational performance of the existing mobile terminal can not only improve the operation speed of the existing algorithms,but also provide implementation spaces for complicated algorithms with larger computation cost and better processing effect.In addition,an auxiliary training module was designed in this paper,for the need of postoperative rehabilitation training for CI implanted patients,which has voice-tospeech conversion function based on speech recognition and speech repeat function.After operation,deafness patients require long-term professional speech rehabilitation training in order to achieve better hearing recovery effect,time-consuming and expensive,which is one of the reasons that many deaf patients give up CI implantation.The auxiliary training module designed in this paper helps recover or reconstruct the speech hearing ability faster and better by text hints and repetitive stimulation,in addition to professional rehabilitation training.Finally,a series of analytical experiments and auditory simulation experiments are designed to verify the reliability of the designed system.