Analysis and Automatic Detection of Aspirated Fricative and Aspirated Nasals

No Thumbnail Available
Date
2022
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Unlike aspiration in stops, aspiration in non-stop consonants is quite rare. Most of the languages that have aspirated non-stop consonants are low-resource languages. Hence, data-driven, quantitative, and statistical analysis of their aspiration phenomena is fairly limited. From the literature review, it has been observed that there is still a need to explore a novel framework that will utilize the advantages of both linguistic and signal- processing knowledge-based approaches for acoustic-phonetic analysis and automatic classi cation of aspirated fricatives and aspirated nasals. To address these issues, we study two phonemes,/s/ and /sh/ in a North-eastern language of India of Tibeto-Burman origin - Rabha, where contrast exists between aspirated and unaspirated counterparts. Also, for the study of aspirated nasals (/m/- /mh/, /n/-/nh/ and /N/-/Nh), we choose the Angami language, another North-eastern language of India of Tibeto-Burman origin. Both languages are low-resourced and are characterized by these unique sounds.
Description
Supervisors: Prasanna, S R M and Sarmah, Priyankoo
Keywords
Fricatives, Nasals, Aspiration, Under-resourced Language, Speech Recognition, Speech Analysis, Perception, Low-resourced Languages, Rabha, Angami, Korean
Citation