Abstract
Introduction
NUBUC (NyU-BU contextually controlled stories Corpus) was developed by New York University, Max Planck Institute for Empirical Aesthetics and Boston University. It contains approximately three hours of English read speech from eight stories focused on linguistic keywords that were created specifically for this corpus, along with transcripts, syntactic annotations and corpus metadata.
Data
Stories are centered on a protagonist and bear a similarity to a modern fairy tale. Each story consists of approximately 2,000 words organized around critical keywords matched along multiple linguistic dimensions. The story texts comprise a total of 1024 sentences and 16,472 words. Sentences across the eight stories have the same number of words, and alternating sentences contain the linguistically equated keyword. Contextual variables are systematically manipulated while holding linguistic and semantic variables constant across sentences and stories.
More information about the story design is included in the documentation. The text of the stories, syntactic annotations, and TextGrid word-aligned transcripts are all UTF-8 encoded.
Each story was read by two different voice actors, one male and one female, in a neutral American English accent. Recordings are 11-12 minutes in duration, for a total of about 90 minutes of continuous speech per speaker. Audio files divided by story as well as by sentence are included in this release; each audio file is presented as a single channel, 11025 Hz, 16-bit, flac compressed wave file.
(2022-05-16)