Related Experiment Videos
Binaural processing model based on contralateral inhibition. III. Dependence on temporal parameters.
J Breebaart1, S van de Par, A Kohlrausch
1IPO, Center for User-System Interaction, Eindhoven, The Netherlands. jeroen.breebaart@philips.com
The Journal of the Acoustical Society of America
|August 25, 2001
Summary
This study presents a computational model of the binaural auditory system. The model successfully predicts human performance in various binaural masking experiments, except for specific conditions involving periodically varying interaural time differences.
Area of Science:
- Auditory Neuroscience
- Computational Auditory Processing
- Psychoacoustics
Background:
- The binaural auditory system processes sound using two ears to enable sound localization and improve speech intelligibility in noise.
- Existing models often struggle to capture the full complexity of human binaural hearing, particularly under dynamic listening conditions.
Purpose of the Study:
- To develop and validate a computational model simulating the signal processing stages of the human binaural auditory system.
- To assess the model's predictive accuracy against psychophysical data from binaural masking experiments.
Main Methods:
- A computational model incorporating monaural and binaural preprocessing stages followed by an optimal detector was developed.
- The model was employed as an artificial observer in simulated three-interval, forced-choice masking experiments.
- Simulations were conducted varying temporal stimulus parameters, including interaural correlation, masker/signal duration, and time-varying interaural differences.
Main Results:
- The model accurately predicted human performance across a range of binaural masking conditions, including varying interaural noise correlation and stimulus durations.
- The model demonstrated predictive capabilities for stimuli with time-varying interaural intensity differences and binaural forward-masking.
- Deviations were observed for periodically varying interaural time differences and for measuring interaural correlation just-noticeable differences as a function of bandwidth.
Conclusions:
- A computational model combining a temporal integrator and an optimal detector effectively accounts for a broad spectrum of binaural masking phenomena.
- The model provides a valuable tool for understanding binaural signal processing but requires further refinement for specific dynamic ITD conditions.