We propose in this paper a method for syllable segmentation based on the Sonority Sequencing Principle, principle by which the sonority inside a syllable increases from its boundaries towards the syllabic nucleus. The sonority function employed was derived from the posterior probabilities of a broad phonetic class recognizer, trained with data coming from an open-source corpus of English stories. We tested our approach on English, Spanish and Catalan and compared the results obtained to those given by an energy-based system. The proposed method outperformed the energy-based system on all three languages, showing a good generalizability to the two unseen languages. We conclude with a discussion of the implications of this work for under-resourced languages.
展开▼
