ABACUS – Demos & Software

ABACUS: Advancing Behavioral and Cognitive Understanding of Speech is a research group that combines AI, linguistics and cognitive science, headed by Bart de Boer.

Home | Members | Projects | Demos & Software

Demos

LeVI: interactive browser demonstration of the LeVI vocal tract model of a learning virtual infant, developed by Heikki Rasilo.
Simulation of language competition: Simulation of the competition between two languages in a city, developed by Peter Dekker. This demo shows how agent-based simulations, models of interactions between individuals, can help give better insight in real-world situations.

Software packages

Parselmouth A Python library for the Praat software, allowing for speech analysis in Python. Developed by Yannick Yadoul. [website] [github]

Code from papers

Agent-based model of conversational priming Code for paper: Dekker, P., Gipper, S. & de Boer, B. (2024). Conversational priming in repetitional responses as a mechanism in language change: evidence from agent-based modelling. Linguistics Vanguard, 10(1), 549-564. [github]
 
Statistical analysis of subject markers in world’s languages. Code for paper: Dekker, P., Gipper, S. & de Boer, B. (2024). 3SG is the most conservative subject marker across languages: An exploratory study of rate of change. In The Evolution of Language: Proceedings of the 15th International Conference (Evolang XV). Madison, WI, USA. [github]
 
Word prediction in computational historical linguistics This is a Jupyter notebook and Python library to demonstrate the use of word prediction using deep learning as an aid in historical linguistics. This notebook accompanies the following paper: Dekker, P., & Zuidema, W. (2021). Word Prediction in Computational Historical Linguistics. Journal of Language Modelling, 8(2), 295–336. The results yielded by this demonstrational notebook may differ somewhat from the results in the article. [github]
 
Agent-based model using Restricted Boltzmann Machines (RBMs) Master thesis project by Joni Muyshondt. [github]
 
Reimplementation of Oudeyer (2005) “The self-organization of speech sounds” Bachelor thesis project by Flynn Steppe. The results of the Oudeyer (2005) are not fully replicated. [github]