---
title: "OpenSubtitles2024 — Tiedemann 2026"
description: "Translation data for conversational, everyday language is scarce, most parallel corpora pair other languages only with English, and MT benchmarks with…"
authors: ["Joerg Tiedemann", "Hengyu Luo"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.700/"
doi: "10.63317/4ivg578ub2ob"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/tiedemann-2026-opensubtitles2024-massively-parallel-dataset-movie/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/tiedemann-2026-opensubtitles2024-massively-parallel-dataset-movie/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2026](https://md.research.elamur.ai/free/papers/2026/page/4/) / OpenSubtitles2024: A Massively Parallel Dataset of Movie Subtitles for MT Development and Evaluation

# OpenSubtitles2024: A Massively Parallel Dataset of Movie Subtitles for MT Development and Evaluation

**Authors:** Joerg Tiedemann, Hengyu Luo · **Year:** 2026 · **Venue:** Proceedings of the Fifteenth Language Resources and Evaluation Conference · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.lrec-1.700/) · [PDF](https://aclanthology.org/2026.lrec-1.700.pdf) · [DOI 10.63317/4ivg578ub2ob](https://doi.org/10.63317/4ivg578ub2ob) · [OpenAlex](https://openalex.org/W7160051162)

**Citations:** 0 citations (openalex-snapshot · 2026-10-14)

## Summary

Translation data for conversational, everyday language is scarce, most parallel corpora pair other languages only with English, and MT benchmarks with wide language coverage are rare and at risk of overuse and contamination. Tiedemann and Luo address these gaps with OpenSubtitles2024, a new release built from user-uploaded subtitles on OpenSubtitles.org that roughly doubles the earlier OpenSubtitles2018.

The pipeline normalizes the subtitles, splits them into sentences, and identifies their language with an improved model. Sentences in different language versions of the same film or episode are then aligned by how much their timestamps overlap. Weak links and mismatched file pairs are filtered out. Badly timed subtitles are re-synchronized using lexical anchor points taken from bilingual dictionaries. Machine-generated subtitles are excluded.

The training data is distributed through OPUS in several formats. A separate held-out collection of recent subtitles serves as a development and test benchmark. It is shared in password-protected packages to reduce contamination and is integrated into GlotEval and LM-Evaluation-Harness. The resource matters for training and evaluating MT and multilingual LLMs, including many translation directions that do not involve English.

## Key points

- Language coverage grows from 63 to 92 languages and variants, among them Ukrainian. The corpus has 7.7 million subtitle files, over 40 billion tokens and about 5.3 billion sentences or fragments, roughly 2 billion sentences more than earlier releases.
- The training portion contains 3,377 bitexts, now the largest parallel corpus in OPUS. It comes as XML with standoff alignments, Moses plain text and TMX, plus monolingual and intra-lingual alignments.
- Alignment quality controls:
  - links with a time-overlap score below 0.5 are dropped;
  - file pairs with a duration ratio under 0.75 are discarded;
  - the best subtitle version for each language pair is chosen by alignment density.
- The bilingual test set covers 2,022 language pairs. A multi-parallel subset aligns 15 subtitles across 40 languages.
- Held-out data is released on GitHub (Helsinki-NLP/OpenSubtitles-devtest) in password-protected form.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/) · [Machine translation](https://md.research.elamur.ai/free/shelves/machine-translation/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{tiedemann2026opensubtitles2024,
  title = {{OpenSubtitles2024: A Massively Parallel Dataset of Movie Subtitles for MT Development and Evaluation}},
  author = {Tiedemann, Joerg and Luo, Hengyu},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/4ivg578ub2ob},
  url = {https://doi.org/10.63317/4ivg578ub2ob}
}
```

- [← previous Open English NameNet: Extending English Wordnet with Names](https://md.research.elamur.ai/free/papers/lex-2026-mccrae-open-english-namenet/)
- [next → ParaCLEAN: Improving Translation Quality through Systematic Parallel Data Cleaning](https://md.research.elamur.ai/free/papers/mash-2026-paraclean-improving-translation-quality-through/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/papers/2026/page/4/)

## References

13 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Simone Balloccu, Patrícia Schmidtová, Mateusz Lango, and Ondrej Dusek. 2024. Leak, cheat, repeat: Data contamination and evaluation malpractices in closed-source LLMs. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 67–93, St. Julian’s, Malta. Association for Computational Linguistics. [↗](https://doi.org/10.48550/arxiv.2402.03927) [our card](https://md.research.elamur.ai/free/papers/balloccu-2024-leak-cheat-repeat-data-contamination/)
2. Leo Gao, Jonathan Tow, Baber Abbasi, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Alain Le Noac’h, Haonan Li, Kyle McDonell, Niklas Muennighoff, Chris Ociepa, Jason Phang, Laria Reynolds, Hailey Schoelkopf, Aviya Skowron, Lintang Sutawika, Eric Tang, Anish Thite, Ben Wang, Kevin Wang, and Andy Zou. 2024. The language model evaluation harness. [find](https://scholar.google.com/scholar?q=Leo+Gao%2C+Jonathan+Tow%2C+Baber+Abbasi%2C+Stella+Biderman%2C+Sid+Black%2C+Anthony+DiPofi%2C+Charles+Foster%2C+Laurence+Golding%2C+Jeffrey+Hsu%2C+Alain+Le+Noac%E2%80%99h%2C+Haonan+Li%2C+Kyle+McDonell%2C+Niklas+Muennighoff%2C+Chris+Oci)
3. Hengyu Luo, Zihao Li, Joseph Attieh, Sawal Devkota, Ona de Gibert, Shaoxiong Ji, Peiqin Lin, Bhavani Sai Praneeth Varma Mantina, Ananda Sreenidhi, Raúl Vázquez, Mengjie Wang, Samea Yusofi, and Jörg Tiedemann. 2025. Gloteval: A test suite for massively multilingual evaluation of large language models. [↗](https://doi.org/10.48550/arxiv.2504.04155)
4. Oscar Sainz, Jon Campos, Iker García-Ferrero, Julen Etxaniz, Oier Lopez de Lacalle, and Eneko Agirre. 2023. NLP evaluation in trouble: On the need to measure LLM data contamination for each benchmark. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 10776–10787, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Oscar+Sainz%2C+Jon+Campos%2C+Iker+Garc%C3%ADa%2DFerrero%2C+Julen+Etxaniz%2C+Oier+Lopez+de+Lacalle%2C+and+Eneko+Agirre%2E+2023%2E+NLP+evaluation+in+trouble%3A+On+the+need+to+measure+LLM+data+contamination+for+each+benchmark%2E) [our card](https://md.research.elamur.ai/free/papers/sainz-2023-nlp-evaluation-trouble-need-measure/)
5. Jörg Tiedemann. 2008. Synchronizing translated movie subtitles. In Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08), Marrakech, Morocco. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=J%C3%B6rg+Tiedemann%2E+2008%2E+Synchronizing+translated+movie+subtitles%2E+In+Proceedings+of+the+Sixth+International+Conference+on+Language+Resources+and+Evaluation+%28LREC%E2%80%9908%29%2C+Marrakech%2C+Morocco%2E+European+Langua)
6. Jörg Tiedemann. 2016. Finding alternative translations in a large corpus of movie subtitle. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), pages 3518–3522, Portorož, Slovenia. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=J%C3%B6rg+Tiedemann%2E+2016%2E+Finding+alternative+translations+in+a+large+corpus+of+movie+subtitle%2E+In+Proceedings+of+the+Tenth+International+Conference+on+Language+Resources+and+Evaluation+%28LREC%E2%80%9916%29%2C+pages+3)
7. Yuemei Xu, Ling Hu, Jiayi Zhao, Zihan Qiu, Kexin Xu, Yuqi Ye, and Hanwen Gu. 2025. A survey on multilingual large language models: Corpora, alignment, and bias. Frontiers of Computer Science, 19(11):1911362. 9. [↗](https://doi.org/10.1007/s11704-024-40579-4)
8. Language Resource References Marta Bañón, Pinzhen Chen, Barry Haddow, Kenneth Heafield, Hieu Hoang, Miquel Esplà-Gomis, Mikel L. Forcada, Amir Kamran, Faheem Kirefu, Philipp Koehn, Sergio Ortiz Rojas, Leopoldo Pla Sempere, Gema Ramírez-Sánchez, Elsa Sarrías, Marek Strelec, Brian Thompson, William Waites, Dion Wiggins, and Jaume Zaragoza. 2020. ParaCrawl: Web-scale acquisition of parallel corpora. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4555–4567, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Language+Resource+References+Marta+Ba%C3%B1%C3%B3n%2C+Pinzhen+Chen%2C+Barry+Haddow%2C+Kenneth+Heafield%2C+Hieu+Hoang%2C+Miquel+Espl%C3%A0%2DGomis%2C+Mikel+L%2E+Forcada%2C+Amir+Kamran%2C+Faheem+Kirefu%2C+Philipp+Koehn%2C+Sergio+Ortiz+Rojas%2C) [our card](https://md.research.elamur.ai/free/papers/banon-2020-paracrawl-web-scale-acquisition-parallel/)
9. Christos Christodouloupoulos and Mark Steedman. 2015. A massively parallel corpus: the bible in 100 languages. Lang. Resour. Eval., 49(2):375–395. [↗](https://doi.org/10.1007/s10579-014-9287-y) [our card](https://md.research.elamur.ai/free/papers/christodouloupoulos-2014-massively-parallel-corpus-bible-100/)
10. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In Proceedings of Machine Translation Summit X: Papers, pages 79–86, Phuket, Thailand. [find](https://scholar.google.com/scholar?q=Philipp+Koehn%2E+2005%2E+Europarl%3A+A+parallel+corpus+for+statistical+machine+translation%2E+In+Proceedings+of+Machine+Translation+Summit+X%3A+Papers%2C+pages+79%E2%80%9386%2C+Phuket%2C+Thailand%2E)
11. Pierre Lison and Jörg Tiedemann. 2016. OpenSubtitles2016: Extracting large parallel corpora from movie and TV subtitles. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), pages 923– 929, Portorož, Slovenia. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Pierre+Lison+and+J%C3%B6rg+Tiedemann%2E+2016%2E+OpenSubtitles2016%3A+Extracting+large+parallel+corpora+from+movie+and+TV+subtitles%2E+In+Proceedings+of+the+Tenth+International+Conference+on+Language+Resources+and+) [our card](https://md.research.elamur.ai/free/papers/lison-2016-opensubtitles2016-extracting-large-parallel-corpora/)
12. Holger Schwenk, Guillaume Wenzek, Sergey Edunov, Edouard Grave, Armand Joulin, and Angela Fan. 2021. CCMatrix: Mining billions of high-quality parallel sentences on the web. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 6490–6500, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Holger+Schwenk%2C+Guillaume+Wenzek%2C+Sergey+Edunov%2C+Edouard+Grave%2C+Armand+Joulin%2C+and+Angela+Fan%2E+2021%2E+CCMatrix%3A+Mining+billions+of+high%2Dquality+parallel+sentences+on+the+web%2E+In+Proceedings+of+the+59th) [our card](https://md.research.elamur.ai/free/papers/schwenk-2019-ccmatrix-mining-billions-high-quality/)
13. Ralf Steinberger, Bruno Pouliquen, Anna Widiger, Camelia Ignat, Tomaž Erjavec, Dan Tufiş, and Dániel Varga. 2006. The JRC-Acquis: A multilingual aligned parallel corpus with 20+ languages. In Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC’06), Genoa, Italy. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Ralf+Steinberger%2C+Bruno+Pouliquen%2C+Anna+Widiger%2C+Camelia+Ignat%2C+Toma%C5%BE+Erjavec%2C+Dan+Tufi%C5%9F%2C+and+D%C3%A1niel+Varga%2E+2006%2E+The+JRC%2DAcquis%3A+A+multilingual+aligned+parallel+corpus+with+20%2B+languages%2E+In+Proceedi) [our card](https://md.research.elamur.ai/free/papers/steinberger-2006-jrc-acquis-multilingual-aligned-parallel/)