Hyper-parameter Adaptation of Conformer ASR Systems for Elderly and Dysarthric Speech Recognition

TL;DR

This paper explores hyper-parameter adaptation for Conformer ASR systems pre-trained on general speech data, improving recognition accuracy for elderly and dysarthric speech through domain-specific tuning.

Contribution

It introduces hyper-parameter adaptation techniques for Conformer ASR models, demonstrating improved performance on elderly and dysarthric speech datasets beyond standard fine-tuning.

Findings

Hyper-parameter adaptation reduces WER by 0.45% on DementiaBank.

Hyper-parameter adaptation reduces WER by 0.67% on UASpeech.

Performance improvements correlate with utterance length ratios.

Abstract

Automatic recognition of disordered and elderly speech remains highly challenging tasks to date due to data scarcity. Parameter fine-tuning is often used to exploit the large quantities of non-aged and healthy speech pre-trained models, while neural architecture hyper-parameters are set using expert knowledge and remain unchanged. This paper investigates hyper-parameter adaptation for Conformer ASR systems that are pre-trained on the Librispeech corpus before being domain adapted to the DementiaBank elderly and UASpeech dysarthric speech datasets. Experimental results suggest that hyper-parameter adaptation produced word error rate (WER) reductions of 0.45% and 0.67% over parameter-only fine-tuning on DBank and UASpeech tasks respectively. An intuitive correlation is found between the performance improvements by hyper-parameter domain adaptation and the relative utterance length ratio between the source and target domain data.