A Probabilistic Modeling Approach to In-situ Trainable Gesture Recognition
Gesture recognition enables a natural extension of the way we currently interact with devices. Commercially available gesture recognition systems are usually pre-trained. We propose a method that allows users to define their own gestures using only a few training examples.
Bayesian Pure-tone Audiometry
We develop a fully probabilistic approach to pure-tone audiometry. By assuming a Gaussian process based response model for test tones, the hearing threshold estimation problem becomes one of Bayesian inference. This allows the use of information-theoretic criteria to select optimal test tones.
Probabilistic Hearing Loss Compensation
Hearing Aid (HA) algorithms need to be tuned (fitted) to match the impairment of each specific patient. The lack of a fundamental HA fitting theory is a strong contributing factor to an unsatisfying sound experience for about 20% of hearing aid patients. We propose here a probabilistic modeling approach to the design of HA algorithms. Our method relies on a generative probabilistic model for the hearing loss problem and provides for automated inference of the corresponding (1) signal processing algorithm, (2) the fitting solution as well as a principled (3) performance evaluation metric. All three tasks are realized as message passing algorithms in a factor graph representation of the generative model, which in principle allows for fast implementation on hearing aid or mobile device hardware. The methods are theoretically worked out and simulated with a custom-built factor graph toolbox for a specific hearing loss model.
A Probabilistic Modeling Approach to Hearing Loss Compensation
The graphical brain: belief propagation and active inference
A Gaussian process mixture prior for hearing loss modeling
Probabilistic Inference-based Reinforcement Learning
Our Approach to Solving Difficult Design Problems
How do you solve a problem that is only vaguely described? We describe here an engineering approach that guides our research on solving vaguely defined problems such as hearing impairment.
Our Mission: Bayesian Intelligent Agents
The primary mission of the BIASlab team is to develop in-situ trainable Bayesian Intelligent Agents for applications to wearable technology.
Bert de Vries
Professor, TU Eindhoven
Tjalling Tjalkens
Associate Professor, TU Eindhoven
Anouk van Diepen
PhD candidate, TU Eindhoven
Thijs van de Laar
PhD candidate, TU Eindhoven
Marco Cox
PhD candidate, TU Eindhoven
Ivan Bocharov
PhD candidate, TU Eindhoven
Quan (Eric) Nguyen
PhD Candidate, TU Eindhoven
Ismail Senoz
Graduate student, TU Eindhoven
Joris Kraak
Senior Software Engineer, GN Hearing
We gratefully acknowledge financial support from our sponsors:
