Method, apparatus and system for neural network hearing aid
The disclosure generally relates to a method, system and apparatus to improve a user's understanding of speech in real-time conversations by processing the audio through a neural network contained in a hearing device. The hearing device may be a headphone or hearing aid. In one embodiment, the disclosure relates to an apparatus to enhance incoming audio signal. The apparatus includes a controller to receive an incoming signal and provide a controller output signal; a neural network engine (NNE) circuitry in communication with the controller, the NNE circuitry activatable by the controller, the NNE circuitry configured to generate an NNE output signal from the controller output signal; and a digital signal processing (DSP) circuitry to receive one or more of controller output signal or the NNE circuitry output signal to thereby generate a processed signal; wherein the controller determines a processing path of the controller output signal through one of the DSP or the NNE circuitries as a function of one or more of predefined parameters, incoming signal characteristics and NNE circuitry feedback.
1 . An ear-worn device, comprising:
front-end circuitry configured to generate audio signals;
communication circuitry configured to receive at least one user input from an external device in wireless communication with the ear-worn device,
wherein the at least one user input comprises a user preference for speech amplification or a user preference for noise reduction;
digital signal processing (DSP) circuitry configured to:
process the audio signals at first time intervals; and
perform audio enhancement on the audio signals based on the at least one user input; and
neural network engine (NNE) circuitry configured to process the audio signals at second time intervals that are less frequent than the first time intervals; and
provide supportive computation for the audio enhancement performed by the DSP circuitry.
2 . A system comprising the ear-worn device of claim 1 and the external device, wherein the external device is configured to display a graphical user interface (GUI) comprising a first GUI element corresponding to the user preference for speech amplification and a second GUI element corresponding to the user preference for noise reduction.
3 . The ear-worn device of claim 1 , wherein the supportive computation comprises detecting speech.
4 . The ear-worn device of claim 1 , wherein the front-end circuitry and the DSP circuitry are in a signal path and the NNE circuitry is not in the signal path.
5 . The ear-worn device of claim 1 , wherein the ear-worn device is configured to characterize an acoustic environment when engaging the NNE circuitry.
6 . The ear-worn device of claim 1 , further comprising a controller configured to engage the NNE circuitry based on detecting a tap on the ear-worn device.
7 . The ear-worn device of claim 1 , further comprising a controller configured to automatically engage the NNE circuitry based on user-agnostic criteria.
8 . The ear-worn device of claim 1 , wherein the NNE circuitry is configured to perform at least 1 billion operations per second.
9 . The ear-worn device of claim 1 , wherein the NNE circuitry is configured to achieve at least 2 billion operations per milliwatt.
10 . The ear-worn device of claim 3 , wherein the NNE circuitry is configured to detect presence of the speech with an associated power consumption of about 2 milliwatts or less.
11 . The ear-worn device of claim 1 , wherein the NNE circuitry is implemented on a single chip in the ear-worn device.