Hearing Device With Neural Network-based Microphone Signal Processing
A hearing system performs nonlinear processing of signals received from a plurality of microphones using a neural network to enhance a target signal in a noisy environment. In various embodiments, the neural network can be trained to improve a signal-to-noise ratio without causing substantial distortion of the target signal. An example of the target sound includes speech, and the neural network is used to improve speech intelligibility.
Claim CLM-00001. 1. A method for transmitting a target sound to a listener, comprising:
comprising: receiving microphone signals including the target sound from microphones; processing the microphone signals using a neural network to produce a processed signal including the target sound, the neural network including a linear signal processing path and a nonlinear signal processing path; producing an output signal including the target sound using the processed signal; and producing an output sound based on the output signal using a speaker.
Claim CLM-00011. 11. A hearing system for transmitting a target sound to a listener, comprising:
microphones configured to receive input sounds including the target sound and to produce microphone signals using the input sounds; a control circuit configured to produce an output signal using a processed signal and including a neural network, the output signal including the target sound, the neural network including a linear signal processing path and a non-linear signal processing path and configured to produce the processed signal by processing the microphone signal through the linear and non-linear signal processing paths; and a speaker configured to produce an output sound using the output signal.