African News Corpus

This consist of a monolingual news corpus for 19 languages from various sources like VOA, BBC, isolezwe etc. - The BBC corpus (except for Yoruba) was extracted from the AfriBERTa corpus, please cite the AfriBERTa paper if you use it. A big thank you to Kelechi Ogueji for providing this corpus - The VOA corpus was extracted from the MOT corpus, please cite the MOT paper if you use it. - The Isolezwe (xho, zul) was crawled as part of the Lacuna NER/POS project with Masakhane, please cite the MAFAND paper for that. - The nya data was part of the AI4D paper. - We thank Jonathan Mukiibi for providing the lug news corpus. - If you use the corpus for amh, hau, ibo, kin, lug, luo, pcm, swa, wol, yor, please cite our MAFT paper. We provide a description of the sources in the paper.