Skip to main content

Intonation Template Matching for Syllable-Level Prosody Encoding.

Rehman, A., Zhang, J. J. and Yang, X., 2023. Intonation Template Matching for Syllable-Level Prosody Encoding. In: Villagrá, P. L. and Li, X, eds. Proceedings of the International Workshop on Cognitive AI 2023 co-located with the 3rd International Conference on Learning & Reasoning (IJCLR 2023). Aachen: CEUR-WS.

Full text available as:

IJCLR2023_paper_31.pdf - Published Version
Available under License Creative Commons Attribution.


Official URL:


We address the challenge of machine interpretation of subtle speech intonations that convey complex meanings. We assume that emotions and interrogative statements follow regular prosodic patterns, allowing us to create an unsupervised intonation template dictionary. These templates can then serve as encoding mechanisms for higher-level labels. We use piecewise interpolation of syllable-level formant features to create intonation templates and evaluate their effectiveness on three speech emotion recognition datasets and declarative-interrogative utterances. The results indicate that individual syllables can be detected for basic emotions with nearly double the accuracy of chance. Additionally, certain intonation templates exhibit a correlation with interrogative implications.

Item Type:Book Section
Additional Information:Bari, Italy, 13-15 November 2023.
Uncontrolled Keywords:intonations; speech processing; emotion recognition; computational paralinguistics
Group:Faculty of Media & Communication
ID Code:39773
Deposited By: Symplectic RT2
Deposited On:01 May 2024 07:22
Last Modified:01 May 2024 07:22


Downloads per month over past year

More statistics for this item...
Repository Staff Only -