ChatSubs: A dataset of movie dialogues in Spanish, Catalan, Basque and Galician

  1. Kharitonova, Ksenia 1
  2. Callejas, Zoraida 2
  3. Pérez-Fernández, David 3
  4. Gutiérrez-Fandiño, Asier 4
  5. Griol, David 1
  1. 1 Department of Software Engineering, University of Granada, Granada, Spain
  2. 2 Department of Software Engineering and Research Centre for Information and Communication Technologies (CITIC-UGR), University of Granada, Granada, Spain
  3. 3 Universidad Autónoma de Madrid, Madrid, Spain
  4. 4 LHF Labs, Bilbao, Spain
Show affiliations +

Editor: Zenodo

Year of publication: 2023

Type: Dataset

DOI: 10.5281/ZENODO.8192331 lock_openOpen access editor

Abstract

<strong>Description</strong>: The ChatSubs dataset contains dialogues in Spanish and three co-official languages of Spain (Catalan, Basque, and Galician). It was obtained from OpenSubtitles and processed to generate clearly segmented dialogues and turns. The dataset consists of 206,706 JSON files, with over 20 million dialogues and 96 million turns, making it one of the largest dialogue corpora available. It serves as an excellent resource for research teams interested in training dialogue models in Spanish, Catalan, Basque, and Galician. <strong>License</strong>: CC BY-NC 4.0.