- Keywords: multilinguality, bitext mining, neural MT, Wikipedia, low-resource languages, joint sentence representation
- TL;DR: Large-scale bitext extraction from Wikipedia: 1620 language pairs in 85 languages, 135M parallel sentences, Systematic NMT evaluation on TED test set.
- Abstract: We present an approach based on multilingual sentence embeddings to automatically extract parallel sentences from the content of Wikipedia articles in 85 languages, including several dialects or low-resource languages. We do not limit the extraction process to alignments with English, but systematically consider all possible language pairs. In total, we are able to extract 135M parallel sentences for 1620 different language pairs, out of which only 34M are aligned with English. This corpus of parallel sentences is freely available (URL anonymized) To get an indication on the quality of the extracted bitexts, we train neural MT baseline systems on the mined data only for 1886 languages pairs, and evaluate them on the TED corpus, achieving strong BLEU scores for many language pairs. The WikiMatrix bitexts seem to be particularly interesting to train MT systems between distant languages without the need to pivot through English.