README2: how to use - ksugi/german_textsegmentation GitHub Wiki

  1. The input text is expected to be a plain text (utf-8).
  • To get the text segmentation more accurate, it is recommended to put line breaks where paragraphs, chapters and texts end. Freestanding lines like title and subtitle are also recommended to put line breaks in your input file.