End to End Deep Neural Network Frequency Demodulation of Speech Signals

TL;DR

This paper introduces an end-to-end deep neural network approach for frequency demodulation of speech signals in FM radio, leveraging prior speech information to improve robustness against noise and disturbances.

Contribution

It presents a novel SDR receiver that uses deep learning for FM demodulation, outperforming traditional methods especially at low SNR conditions.

Findings

High performance detection in noisy environments

Outperforms established methods in mean square error

Improves perceptual speech quality scores

Abstract

Frequency modulation (FM) is a form of radio broadcasting which is widely used nowadays and has been for almost a century. We suggest a software-defined-radio (SDR) receiver for FM demodulation that adopts an end-to-end learning based approach and utilizes the prior information of transmitted speech message in the demodulation process. The receiver detects and enhances speech from the in-phase and quadrature components of its base band version. The new system yields high performance detection for both acoustical disturbances, and communication channel noise and is foreseen to out-perform the established methods for low signal to noise ratio (SNR) conditions in both mean square error and in perceptual evaluation of speech quality score.