5 ms·

Meta AI open-sources NLLB-200 model that translates 200 languages

• pagekicker 4y ago
  Trying to figure out the most expedient way to set it up and test it. Anyone have any ideas?
  • whitten 4y ago
    Perhaps a docker container? If you set one up, I would like to know how it should be done
• pineaux 4y ago
  Is there a place where I can test this online?
  • tim-- 4y ago
    You can test it over at Wikipedia: https://en.wikipedia.org/wiki/Special:ContentTranslation https://en.wikipedia.org/wiki/Special:ContentTranslation
  • mdaniel 4y ago
    Their more "marketing" page <https://about.fb.com/news/2022/07/new-meta-ai-model-translates-200-languages-making-technology-more-accessible/ https://about.fb.com/news/2022/07/new-meta-ai-model-translat...> linked to a "try it here": https://nllb.metademolab.com/ https://nllb.metademolab.com/
• daedalus2027 4y ago
  I can not help but wonder if it can translate any non previously seen language like protoindoeuropean.
• mdaniel 4y ago
  (a) it does not appear to be "open source" for any such definition (b) the paragraph that seems to be actionable has no hyperlinks for ether word "here" > You can explore a demo of NLLB-200 here, showing how the model can translate stories from around the world, and read the research paper here.
  • charcircuit 4y ago
    >it does not appear to be "open source" for any such definition The code for it is licensed under MIT and the pretrained model was released under CC-BY-NC 4.0.
  • code51 4y ago
    Here you go: https://github.com/facebookresearch/fairseq/tree/nllb https://github.com/facebookresearch/fairseq/tree/nllb
• solarkraft 4y ago
  That's nice. Translation is one of those fields that, besides maybe LibreTranslate, free software was always quite far behind on. Tangent: Why does free software generally lack machine learning features? Is the expertise just not there or is it rather the cultural stigma of "evil black-boxes"?