Appearance
Acknowledgements
The Diction keyboard corrects and suggests words with word lists and word-pair statistics computed from large text collections. The keyboard data holds words and counts only. No sentence or passage from any collection is included.
These collections ask for credit:
- Leipzig Corpora Collection, Universität Leipzig, wortschatz.uni-leipzig.de. D. Goldhahn, T. Eckart and U. Quasthoff (2012), LREC 2012. Corpora for download are licensed CC BY.
- OpenSubtitles2024, OPUS, opus.nlpl.eu. J. Tiedemann and H. Luo (2026), LREC 2026. Subtitles from www.opensubtitles.org. Contains information from OpenSubtitles2024, which is made available under the Open Data Commons Attribution License (ODC-By) v1.0.
- Wikipedia, by Wikipedia contributors, wikipedia.org. Licensed CC BY-SA.
- ParlaMint 5.0, T. Erjavec et al., CLARIN.SI, hdl.handle.net/11356/2004. Licensed CC BY 4.0.
- WordReference forum corpus, A. Berdicevskis (2020), Språkbanken Text, doi.org/10.23695/a5k7-az71. Licensed CC BY 4.0.
- Google Books Ngram, Google, books.google.com/ngrams. Licensed CC BY 3.0.
Changes
Word and word-pair frequencies were counted, filtered, thresholded and converted into the keyboard's own file formats. Each collection is provided by its authors as is, without warranties. See its licence for the details.
Licence of the keyboard data
The English and Italian keyboard data are licensed under CC BY-SA 4.0, because they are built in part from Wikipedia. All other keyboard language data is © Diction Labs, all rights reserved.
Questions go to hello@diction.one.