Method and system for performing sound scene recognition on artificial cochlea by means of mobile equipment

Abstract

The invention discloses a method and a system for sound scene recognition on an artificial cochlea by means of mobile equipment. The method mainly comprises the following steps: enabling a DSP moduleto transmit a received sound signal to a central processing unit through a low-power-consumption Bluetooth technology after the central processing unit receives a scene recognition abnormality signal;enabling the central processing unit to start to carry out sound scene recognition depth calculation according to the sound signal, and after calculation of a plurality of frames of the sound signal,enabling the central processing unit to obtain new scene recognition information and send the new scene recognition information to the DSP module; and enabling the DSP module to compare the new sceneidentification information with the current scene identification information stored in a register, if the scenes are judged to be different, changing a sound signal processing strategy correspondingto the new scene, and recovering eigenvalue calculation again. According to the invention, by utilization of a low-power-consumption wireless sound transmission technology of a speech processor and the higher computing power of the mobile equipment, the reliability of sound scene recognition is improved; and through distributed computing, the use time and power consumption of equipment resources are reduced.

Description

technical field [0001] The invention belongs to the field of medical devices, and in particular relates to a method and system for cochlear implants to recognize sound scenes by means of mobile devices. Background technique [0002] With the continuous development of cochlear implant technology, it has gradually become a mainstream application to use different strategies to process sound signals in different scenarios, so that users can obtain the best listening experience, and further improve the speech recognition rate of users in different scenarios. [0003] In the actual application process, how to perform scene recognition stably and reliably has become a bottleneck problem that limits the development of this technology. For the cochlear implant speech processor, due to the limitation of power consumption, its processing speed is limited, but the sound signal is always incoming in real time, and it is necessary to complete the sound signal frequency domain calculation,...

AI Technical Summary

Problems solved by technology

Therefore, the computational burden of sound scene recognition is a heavy burden on the cochlear implant speech processor

On the other hand, the accuracy of sound scene recognition is closely related to the amount of calculation. Current research points to complex machine learning or neural network algorithms for sound scene recognition. In the actual design process, it is necessary to repeatedly weigh the amount of calculation and accuracy. It is difficult to find an algorithm that takes into account both aspects

Images