Open Collections

A model for auditory lateralization in non-stationary multi-source environments

University of British Columbia

The problem addressed in this thesis is the determination of the positions of sound sources in non-stationary multi-source environments. This problem is approached by developing models that mimic the processing of sounds by the auditory system. It is well known that in the localization process the auditory system utilizes interaural intensity and time differences (lID and ITD) and the interaural envelope delay (lED). However, the way such cues are estimated and organized by the auditory system in non-stationary multi-source situations is not known. It is argued in this thesis that the auditory localization process can be divided into three serial processing stages: decomposition, localization, and integration (DLI). Specifically, the signals detected by the two ears are first decomposed into their spectro-temporal distributions as represented in the neural activities of the auditory nerve fibers. Short-time spatial attributes, in terms of the localization cues, are then determined from energy concentrations in these distributions. A spatial scene of acoustic events is finally built by integrating the energy concentrations according to their spatial attributes. A unique DLI model is proposed in which short-time cue estimation is realized by a process of pattern recognition and comparison using neural networks, and the spatial scene is represented by short-time cue distributions. Three implementations of the DLI model, which model separate auditory pathways responsible for three different types of cue sensitivity (11D, lTD. and lED) observed in the auditory system, are developed, and their performance in estimating short-time cue distributions are studied by computer simulation. It is shown that there are unexplored patterns in the neural signals carried by the auditory nerve fibers that are important for auditory localization. These patterns are shown to contain good indications of interaural differences, and can be used to obtain robust short-time cue estimates by training neural networks that have relatively simple structures. Furthermore, while such networks can be trained using the simplest types of inputs, they show the ability to generalize and perform well with more complex stimuli. It is demonstrated that the model works well in noise and in non-stationary multi-source situations. The same model structure can be trained to estimate different localization cues, suggesting that the underlying structure of the different pathways responsible for different types of cue sensitivities in the auditory system may not necessarily be different. The receptive connection patterns of the hidden neurons in the model indicate that the spectro-temporal response properties of binaural neurons in the auditory system may play an important role in auditory localization, and that excitatory and inhibitory inputs to the binaural neurons play equally important roles in localization.

Thesis/Dissertation

application/pdf

2009-04-22

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/7503

Electrical and Computer Engineering

University of British Columbia

UBCV

DSpace