Defence of dissertation in the field of Speech and Language Technology, M.Sc. (Tech.) Heikki Rasilo
The title of thesis is “Computational models of associative learning in the acquisition of speech imitation, acoustic word models and word-meaning mappings”
How do children learn to imitate speech of other people, extract frequent word patterns from continuous speech, and assign meanings to words spoken by others? Children live in a confusing and ambiguous environment, where learning of these phenomena requires long-term exposure to speech and related objects or contexts, and effective deductive skills in the brain. This thesis studies three language learning phenomena, and provides candidate cognitive processing mechanisms that make language learning possible with human-like language input. The discovered mechanisms may help to understand real cognitive processes in the human brain, or provide practical learning methods for machine learning applications.
Opponent: Professor Roger K. Moore, University of Sheffield, Great Britain
Custos: Professor emeritus Unto Laine, Aalto University School of Electrical Engineering, Department of Signal Processing and Acoustics.
Supervisors: Professor Paavo Alku and Professor emeritus Unto Laine.
Contact information: Heikki Rasilo, +32483047487, firstname.lastname@example.org