Repository files navigation

greek_scansion is designed to automatically annotate Ancient Greek hexameter. That is, it tries, for any valid hexameter verse, to find the correct sequence of long and short syllables.

The repository contains the program code and two evaluation scripts. Further information on both is provided in the appropriate folders. English readers will find a description of the approach and algorithm in thispaper. There also is a related Arxiv paper.

My knowledge of Ancient Greek is very limited. If you are a philologist or have expert knowledge on Ancient Greek and know how to improve the linguistic parts of the programme, please, do one of the following:

• open an issue detailing the problem and your improvement,
• or fork the repo, change the code, and make a pull request.

Anne-Kathrin Schumann, Christoph Beierle, Norbert Blößner: "Using finite-state machines to automatically scan Ancient Greek hexameter". Digital Scholarship in the Humanities 37(1), pp.242-253.PDF.

The project is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike license.

The project has been developed and tested on a Windows 10 machine, using Python 3.6.4.

The finite-state automata have been implemented using thetransitions module by Alexander Neumann.

The finite-state transducer uses theHelsinki Finite-State Tools.

I exported my Greek verses from a heritage MySQL database, using aMySQL Python Connector. If you have a similar requirement, you might consider trying this. Otherwise, you obviously don't need to bother.

I also tried out thesyllabification module by James Tauber. In the end, I didn't use it for my annotations, so if you don't want to install it, you can get rid of this part of the code.

The baseline script references theClassical Language Toolkit (CLTK). On my machine, I managed to run this only from the Ubuntu subsystem. AND it actually does not properly annotate hexameter verses. So this dependency is very much an optional one.

A reasonable baseline can be got fromHope Ranker's hexameter library.