SimonXIX’s avatarSimonXIX’s Twitter Archive—№ 44,426

  1. Installing an OCR engine for multiple languages so that Soas' digital library can automatically OCR images uploaded.
    On Internet Archive Internet Archive logo SimonXIX’s avatar ❤️ 3 Favorites
    1. …in reply to @SimonXIX
      Currently testing with Arabic and... it needs some fine-tuning.
      Permalink On Internet Archive Internet Archive logo SimonXIX’s avatar ❤️ 1 Favorite
      1. …in reply to @SimonXIX
        Left: the Arabic image I tried to OCR. Right: the resulting OCR text.
        oh my god twitter doesn’t include alt text from images in their APIoh my god twitter doesn’t include alt text from images in their API
        Permalink On Internet Archive Internet Archive logo SimonXIX’s avatar
        1. …in reply to @SimonXIX
          The top line is spot-on. The bottom line - the one with more diacritics - is not.
          Permalink On Internet Archive Internet Archive logo SimonXIX’s avatar