About this project

This repository contains finite state transducer (FST) source files and Constraint Grammar resources for the Tsuut'ina language, also known as Sarsi. It is part of the GiellaLT project, an open-source multi-lingual infrastructure for building language resources from scratch. The project provides morphological analysers, proofing tools (spell checkers), and dictionaries for the Tsuut'ina language. The data and implementation are licensed under the terms specified in the LICENSE file. Key components include: - Finite state source files for morphology - Constraint Grammar rules for disambiguation - Speller files in .bhfst format for desktop and mobile platforms - Dictionary resources To build and use the tools, you need an FST compiler such as HFST, Foma, or Xerox Xfst, along with VislCG3 Constraint Grammar tools. Installation instructions are provided for macOS, Ubuntu/Debian, and RedHat/Fedora systems. The project follows the standard GNU build system: - ./autogen.sh to clone dependencies - ./configure to configure the build - make to compile - make install to install (requires root) Nightly and CI/CD installation packages are available for Windows, macOS, and mobile platforms, though these are not tested for language quality and may contain regressions. Proofing tools and keyboards can be installed using the Divvun Installer. Speller files can be tested with divvunspell. Documentation is available at the GiellaLT website and the project-specific documentation page. If you use language data from multiple GiellaLT languages, consider citing the LREC 2022 article by Wiechetek et al. on building open-source multi-lingual infrastructure.