An objective evaluation of Hearing Aids and DNN-based speech enhancement in complex acoustic scenes

TL;DR

This study objectively compares high-end hearing aids and DNN-based speech enhancement algorithms in complex acoustic scenes, revealing that DNN methods outperform traditional HA algorithms in noise suppression and intelligibility.

Contribution

The paper introduces a novel evaluation framework using binaural datasets and realistic noise scenarios to compare hearing aids and DNN-based speech enhancement.

Findings

DNN-based enhancement outperforms HA algorithms in noise suppression

DNN methods improve objective speech intelligibility metrics

Evaluation uses a binaural dataset and realistic acoustic environments

Abstract

We investigate the objective performance of five high-end commercially available Hearing Aid (HA) devices compared to DNN-based speech enhancement algorithms in complex acoustic environments. To this end, we measure the HRTFs of a single HA device to synthesize a binaural dataset for training two state-of-the-art causal and non-causal DNN enhancement models. We then generate an evaluation set of realistic speech-in-noise situations using an Ambisonics loudspeaker setup and record with a KU100 dummy head wearing each of the HA devices, both with and without the conventional HA algorithms, applying the DNN enhancers to the latter. We find that the DNN-based enhancement outperforms the HA algorithms in terms of noise suppression and objective intelligibility metrics.