A large collection of text data spanning multiple languages, often created by applying a single processing pipeline across diverse language communities.