A Fully Convolutional Neural Network Approach to End-to-End Speech Enhancement

TL;DR

This paper introduces a fully convolutional neural network for end-to-end speech enhancement, capable of isolating a target speaker's voice from noisy backgrounds, with potential applications in hearing aids.

Contribution

It presents a novel FCN-based approach that learns speaker-specific models and generalizes well to new speakers and noise conditions.

Findings

Effective separation of target speech from noise

Generalizes to unseen speakers and environments

Robust performance across different SNR levels

Abstract

This paper will describe a novel approach to the cocktail party problem that relies on a fully convolutional neural network (FCN) architecture. The FCN takes noisy audio data as input and performs nonlinear, filtering operations to produce clean audio data of the target speech at the output. Our method learns a model for one specific speaker, and is then able to extract that speakers voice from babble background noise. Results from experimentation indicate the ability to generalize to new speakers and robustness to new noise environments of varying signal-to-noise ratios. A potential application of this method would be for use in hearing aids. A pre-trained model could be quickly fine tuned for an individuals family members and close friends, and deployed onto a hearing aid to assist listeners in noisy environments.