Speech Prosody 2022
23-26 May 2022
Lisbon, Portugal
Digital prosodic typology: Machine learning in prosodic typological analysis
Organizers:
Daniil Kocharov (Saint Petersburg State University)
Juraj Šimko (University of Helsinki)
Priyankoo Sarmah (Indian Institute of Technology in Guwahati)
The recent development of digital data processing techniques, including powerful machine learning approaches, have opened completely new avenues for addressing old and difficult scientific questions formulated within linguistics. These questions include complex diachronic and synchronic aspects of linguistic interactions leading to the rich variety of the world's languages and typological relationships among them, in particular in prosody.
We encourage submissions of scientific papers focusing on machine learning based comparison of prosodic patterns in different languages, dialects or varieties. We are particularly interested in novel approaches to interpretability of the results obtained by deep learning systems, data mining methods and other machine learning techniques in terms of theoretical prosodic typology. We strongly believe that this special session can pave the way towards creating common standards and technologies, as well as shared corpora, that will benefit the speech prosody research community.
@2020. All rights reserved. Webmaster: Ricardo Sousa & Sónia Frota