Creating a Spanish Speech Corpus to Develop Digital Dementia Biomarkers Using Machine Learning

Abstract

Dementia is one of the most prevalent diseases affecting older adults in Mexico. There has been increasing interest in the development of digital biomarkers of dementia based on the analysis of speech. The availability of high-quality speech corpus is important to advance this line of research. However, there are no publicly available dataset in Spanish for this purpose. Therefore, we describe a protocol to capture Spanish audio from older adults for dementia research. We describe the lessons learned and adjustments to the protocol that emerged from a pilot study.