Studying Luxembourgish Phonetics via Multilingual Forced Alignments

Luxembourgish, a Germanic-Franconian language, is embedded in a multilingual context on the divide between Romance and Germanic cultures and remains one of Europe’s under-described languages. This paper investigates the similarity between Luxembourgish phone segments with German, French and English via forced speech alignment techniques. Making use of monolingual acoustic seed models from these three languages, as well as “multilingual” models trained on pooled speech data we investigated whether Luxembourgish was globally better represented by one of the individual languages or by the multilingual model. Although French words are often interspersed in spoken Luxembourgish, forced alignments show a clear preference for Germanic acoustic models, with only a limited usage of the French ones. While globally, the German models provide the best match, a phonebased analysis, shows language-specific preferences: French is preferred for rounded front vowels, nasal vowels and // whereas English is more frequently used for diphthongs. The proposed method enables the acoustic match between phonemes in different languages to be quantified and opens new perspectives in language processing studies for low e-resourced languages and for L2 learning.