Noise suppression system using deep learning for smart devices
Koki Takenaka, Kenji Ozawa · 2022 IEEE 4th Global Conference on Life Sciences and Technologies (LifeTech) · 2022
Noise suppression is beneficial for improving the operational accuracy of voice-controlled devices. In this paper, we propose a noise suppression system with a 14-cm long microphone array that is designed to be mounted on a device. Output signals from the array are regarded as a two-dimensional (2D) image and its 2D spectrum is obtained. Our system uses a deep neural network to estimate the amplitude and phase of noise signals in the spectrum, and extracts the target signal by subtracting the estimated noise signal from the input signal. A computer simulation experiment showed 20-dB suppression for two noise sources.