Skip to content

Fonts for Tesseract training

Stefan Weil edited this page Nov 5, 2017 · 6 revisions

Fonts for Tesseract training

See also https://github.com/tesseract-ocr/tesseract/wiki/Fonts.

Here we collect information on freely available fonts which can be used to train Tesseract. Our focus is on fonts used in European printed books of 16th to 19th century, for example Fraktur fonts or Antiqua fonts with special characters (like the long S) and ligatures used in those books.

Debian packages

tbd. = more information still missing

Fonts for old European texts

  • fonts-blankenburg – Blankenburg, Fraktur, 100 %, ligatures
  • fonts-oldstandard – tbd.
  • fonts-yanone-kaffeesatz – tbd.
  • ttf-adf-berenis – klassizistische Antiqua, tbd.
  • maybe more from the ttf-adf-* fonts

Software

  • fontforge – tbd.

Other fonts

  • IM FELL – Antiqua, 100 %, ligatures
  • Old Standard

More links to font sources

Script examples

Other links