Suomeksi

Researcher of the Month: Sofoklis Kakouros

Sofoklis Kakouros
Photo: Sofoklis Kakouros

Kielipankki – The Language Bank of Finland offers a comprehensive set of resources, tools and services in a high-performance environment. Sofoklis Kakouros tells us about his research on prosody and its associated phenomena.

Who are you?

I am Sofoklis Kakouros, a postdoctoral researcher with the Phonetics and Speech Synthesis Research Group in the Department of Digital Humanities at the University of Helsinki. Before joining this group, I held research positions at different universities across Finland and in the Netherlands, and I also worked in the industry as a speech scientist. My background centers on signal processing, cognitive science, and phonetics.

What is your research topic?

My research interests are rooted in speech and language, with a particular emphasis on understanding prosody and its associated phenomena. Prosody is about how we say something rather than what we say; it adds meaning beyond the words themselves. This includes elements like intonation and timing. Over the years, I have explored various aspects of prosody, focusing on information-theoretic processes within this domain. Overall, my work enhances our comprehension of how acoustic and linguistic variations are statistically organized into the prosody we perceive. For the past years, I have been working in the Research Council of Finland project titled ”Computational Modeling of Prosody in Speech”, aiming to understand the statistical organization in speech acoustics and its connections to prosodic dimensions such as prominence and emotions. This research has numerous applications, including the prosodic analysis of dialectal varieties and parliamentary speech.

How is your research related to Kielipankki?

To effectively analyze and train computational models for speech, an increasing amount of data is required. Kielipankki offers a diverse platform that provides access to the essential resources needed for my research, including materials for speech and language studies. In a recent project conducted by our group, I utilized the Finnish ASR corpus from Kielipankki to analyze recordings of Finnish parliamentary speeches.

Recent publications

Vainio, M., Suni, A., Šimko, J., and Kakouros, S. (2024). The Power of Prosody and Prosody of Power: An Acoustic Analysis of Finnish Parliamentary Speech. In Proceedings of the Conference of the Speech Prosody Special Interest Group (SProSIG) of the International Speech Communication Association – Speech Prosody (SpeechPro-2024), Leiden, The Netherlands, pp. 662–666. 10.21437/SpeechProsody.2024-134

Kakouros, S., Šimko, J., Vainio, M., and Suni, A. (2023). Investigating the Utility of Surprisal from Large Language Models for Speech Synthesis Prosody. In Proceedings of the 12th ISCA Speech Synthesis Workshop (SSW-2023), Grenoble, France, pp. 127–133. 10.21437/SSW.2023-20

Kakouros, S. and O’Mahony, J. (2023). What does BERT learn about prosody? In R. Skarnitzl, & J. Volín (Eds.), Proceedings of the 20th International Congress of Phonetic Sciences (ICPhS-2023) (pp. 1454-1458). GUARANT International spol. s r.o.., Prague, Czechia. https://www.internationalphoneticassociation.org/icphs-proceedings/ICPhS2023/full_papers/622.pdf

Kakouros, S., Stafylakis, T., Mošner, L., and Burget, L. (2023). Speech-based emotion recognition with self-supervised models using attentive channel-wise correlations and label smoothing. In Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP-2023), Rhodes, Greece, pp. 1–5. 10.1109/ICASSP49357.2023.10094673

Corpora

 

The FIN-CLARIN consortium consists of a group of Finnish universities along with CSC – IT Center for Science and the Institute for the Languages of Finland (Kotus). FIN-CLARIN helps the researchers of Social Sciences and Humanities to use, refine, preserve and share their language resources. The Language Bank of Finland is the collection of services that provides the language materials and tools for the research community.

All previously published Language Bank researcher interviews are stored in the Researcher of the Month archive. This article is also published on the website of the Faculty of Arts of the University of Helsinki.