parla_text_corpus / README.md
albertvillanova's picture
Fix `license` metadata (#1)
a388264
metadata
annotations_creators:
  - no-annotation
language_creators:
  - various
language:
  - ca
license:
  - cc-by-4.0
multilinguality:
  - monolingual
pretty_name: ParlaTextCorpus
size_categories:
  - 100k<n<1M
source_datasets:
  - found
task_categories:
  - sequence-modeling
task_ids:
  - language-modeling
tags:
  - robust-speech-event

ParlaTextCorpus

Spoken text corpus for Catalan. Derived and cleaned from three sources. OpenSubtitles, Tv3Parla and Festcat.