BEA – A multifunctional Hungarian spoken language database

Abstract

In diverse areas of linguistics, the demand for studying actual language use is on
\nthe increase. The aim of developing a phonetically-based multi-purpose database of
\nHungarian spontaneous speech, dubbed BEA2, is to accumulate a large amount of
\nspontaneous speech of various types together with sentence repetition and reading.
\nPresently, the recorded material of BEA amounts to 260 hours produced by 280
\npresent-day Budapest speakers (ages between 20 and 90, 168 females and 112
\nmales), providing also annotated materials for various types of research and practical
\napplications.