Refining Automatic Speech Recognition System for older adults

TL;DR

This paper develops an improved automatic speech recognition system tailored for seniors aged 80+ with limited training data, using transfer learning and attention mechanisms to enhance performance.

Contribution

It introduces a transfer learning approach combined with attention mechanisms specifically designed to improve ASR accuracy for older adults with minimal data.

Findings

Transfer learning significantly boosts ASR performance for seniors.

Attention mechanisms further improve recognition accuracy.

Achieved 1.58% absolute improvement over baseline transfer learning model.

Abstract

Building a high quality automatic speech recognition (ASR) system with limited training data has been a challenging task particularly for a narrow target population. Open-sourced ASR systems, trained on sufficient data from adults, are susceptible on seniors' speech due to acoustic mismatch between adults and seniors. With 12 hours of training data, we attempt to develop an ASR system for socially isolated seniors (80+ years old) with possible cognitive impairments. We experimentally identify that ASR for the adult population performs poorly on our target population and transfer learning (TL) can boost the system's performance. Standing on the fundamental idea of TL, tuning model parameters, we further improve the system by leveraging an attention mechanism to utilize the model's intermediate information. Our approach achieves 1.58% absolute improvements over the TL model.