VOTE400(Voide Of The Elderly 400 Hours): A Speech Dataset to Study Voice Interface for Elderly-Care

TL;DR

VOTE400 is a comprehensive Korean speech dataset designed to improve voice recognition systems for elderly users, aiding elderly-care robots with more accurate speech interface capabilities.

Contribution

The paper presents a large-scale, elderly-specific Korean speech dataset and demonstrates its effectiveness in enhancing speech recognition accuracy for elderly voices.

Findings

Speech recognition trained on VOTE400 outperforms conventional systems.

The dataset includes 300 hours of dialog and 100 hours of read speech.

The dataset supports elderly-care robot voice interface development.

Abstract

This paper introduces a large-scale Korean speech dataset, called VOTE400, that can be used for analyzing and recognizing voices of the elderly people. The dataset includes about 300 hours of continuous dialog speech and 100 hours of read speech, both recorded by the elderly people aged 65 years or over. A preliminary experiment showed that speech recognition system trained with VOTE400 can outperform conventional systems in speech recognition of elderly people's voice. This work is a multi-organizational effort led by ETRI and MINDs Lab Inc. for the purpose of advancing the speech recognition performance of the elderly-care robots.