HASA-net: A non-intrusive hearing-aid speech assessment network

TL;DR

HASA-Net is a novel deep learning model that non-intrusively assesses speech quality and intelligibility for hearing aid users, considering hearing-loss factors, and correlates well with established intrusive metrics.

Contribution

It introduces the first unified DNN-based non-intrusive assessment model tailored for hearing aids, incorporating hearing-loss patterns.

Findings

High correlation with HASQI and HASPI metrics

Simultaneous assessment of quality and intelligibility

Incorporates hearing-loss factors into evaluation

Abstract

Without the need of a clean reference, non-intrusive speech assessment methods have caught great attention for objective evaluations. Recently, deep neural network (DNN) models have been applied to build non-intrusive speech assessment approaches and confirmed to provide promising performance. However, most DNN-based approaches are designed for normal-hearing listeners without considering hearing-loss factors. In this study, we propose a DNN-based hearing aid speech assessment network (HASA-Net), formed by a bidirectional long short-term memory (BLSTM) model, to predict speech quality and intelligibility scores simultaneously according to input speech signals and specified hearing-loss patterns. To the best of our knowledge, HASA-Net is the first work to incorporate quality and intelligibility assessments utilizing a unified DNN-based non-intrusive model for hearing aids. Experimental results show that the predicted speech quality and intelligibility scores of HASA-Net are highly correlated to two well-known intrusive hearing-aid evaluation metrics, hearing aid speech quality index (HASQI) and hearing aid speech perception index (HASPI), respectively.