R and Praat scripts that facilitate the force-alignment process for minority languages
This repository aims to help the creation of tools for preparing minority languages for forced-alignment. This targets languages to be analysed at the phonetic/phonological level but without an existing model to be used for forced-aligner programs, and mainly oriented towards Montreal Forced Aligner (MFA). These languages must have some degree of documentation, enough to add entries as new words to existing aligner dictionaries.
- Languages with some degree of documentation. Ideally, languages with time aligned transcriptions (e.g. in Elan and Flex) to build a dictionary from.
- At least a rough description of the phonological system