Analysing Iterative Machine Learning Algorithms with Information Geometric Methods. Online machine learning problems arise from situations where data is provided a point at a time. There are many classical algorithms for solving such problems based on the principle of stochastic gradient descent. Recent research by the CIs and others have thrown up interesting but diverse geometric connections that offer new insights. The proposed research aims to integrate the understanding of these algori ....Analysing Iterative Machine Learning Algorithms with Information Geometric Methods. Online machine learning problems arise from situations where data is provided a point at a time. There are many classical algorithms for solving such problems based on the principle of stochastic gradient descent. Recent research by the CIs and others have thrown up interesting but diverse geometric connections that offer new insights. The proposed research aims to integrate the understanding of these algorithms with the aim of designing algorithms better able to exploit prior knowledge, and to extend existing algorithms to new problem domains thus offering well principled and well understood algorithms for solving a variety of novel online problems.Read moreRead less
Breathing and snoring sound analysis in sleep apnea. About 800,000 Australians suffer from the disease sleep Apnoea (OSA) which has snoring as its earliest symptom. We develop electronics and snore processing algorithms to classify snorers into OSA-positive and OSA-negative classes, based on advanced technology derived from speech recognition systems.
Discovery Early Career Researcher Award - Grant ID: DE170100128
Funder
Australian Research Council
Funding Amount
$395,000.00
Summary
Information processing in the brain. This project aims to understand the brain's functional organisation by developing non-invasive methods to characterise connectivity between interacting brain regions. No model-based methods to compute directional coupling between brain regions can be applied to large scale networks for resting state functional MRI data. This capability would be a major breakthrough in neuroimaging, given uninformative (non-directional) network connectivity analysis restricts ....Information processing in the brain. This project aims to understand the brain's functional organisation by developing non-invasive methods to characterise connectivity between interacting brain regions. No model-based methods to compute directional coupling between brain regions can be applied to large scale networks for resting state functional MRI data. This capability would be a major breakthrough in neuroimaging, given uninformative (non-directional) network connectivity analysis restricts research. This project is expected to advance our understanding of information processing in the brain by providing a mechanistic approach to functional integration.Read moreRead less
Electronic Auditory Pathway. We will develop electronic building blocks to investigate biological signal processing. In particular, we will investigate the auditory pathway and develop the most accurate electronic model of the biological cochlea and auditory nerve. These will be followed by electronic circuits that model the processing of sensory signals in the brain. Processing signals with neural spikes offers distinct advantages over current analogue and digital signal processing techniques i ....Electronic Auditory Pathway. We will develop electronic building blocks to investigate biological signal processing. In particular, we will investigate the auditory pathway and develop the most accurate electronic model of the biological cochlea and auditory nerve. These will be followed by electronic circuits that model the processing of sensory signals in the brain. Processing signals with neural spikes offers distinct advantages over current analogue and digital signal processing techniques in terms of noise, energy consumption and extraction of temporal information. We will implement the first spike-based models of pitch and timbre perception, and a neural model of speech recognition in noisy environments.Read moreRead less
Online Structural Health Monitoring (SHM) System Using Active Diagnostic Sensor Network. It is imperative to remain technological leading for Australian research community. But current lack of reliable technique in structural health monitoring in Australia is considerably impeding her competition with other developed countries in areas of forefront technology. Outcomes of the project will lead to an online structural health monitoring system incorporated with active diagnostic sensor network, re ....Online Structural Health Monitoring (SHM) System Using Active Diagnostic Sensor Network. It is imperative to remain technological leading for Australian research community. But current lack of reliable technique in structural health monitoring in Australia is considerably impeding her competition with other developed countries in areas of forefront technology. Outcomes of the project will lead to an online structural health monitoring system incorporated with active diagnostic sensor network, related software and hardware, novel signal processing technique, and artificial intelligence algorithm-based damage identification scheme. Its successful applications in various industries, e.g. aerospace, maritime and civil, are expected to bring significant improvement in operation safety and great benefit in reducing maintenance cost.Read moreRead less
Computational neural modelling of bottom-up information and top-down attention in auditory perception. The aim of this project is to gain a better understanding of the ways in which our auditory cortex functions. This project will make a significant contribution to this important and fundamental aspect of brain science and brain-inspired computation. The outcome will be to build a computational model of the auditory cortex, through simulation of the detailed neuronal responses using spiking neur ....Computational neural modelling of bottom-up information and top-down attention in auditory perception. The aim of this project is to gain a better understanding of the ways in which our auditory cortex functions. This project will make a significant contribution to this important and fundamental aspect of brain science and brain-inspired computation. The outcome will be to build a computational model of the auditory cortex, through simulation of the detailed neuronal responses using spiking neurons. Applications will develop improved processing strategies for automatic speech recognition, hearing aids, bionic ears (cochlear implants), robotics and other machine processing systems.Read moreRead less
Robust speech recognition in realistic hostile environments. Australia leads the world in the adoption of speech recognition technology but sadly lags in the development of the fundamental advances in the area. This research will help propel Australia to the forefront of new innovations in speech recognition technology and contributions to fundamental science. Our project will provide an excellent training ground for graduate students and researchers, with the real possibility of significant com ....Robust speech recognition in realistic hostile environments. Australia leads the world in the adoption of speech recognition technology but sadly lags in the development of the fundamental advances in the area. This research will help propel Australia to the forefront of new innovations in speech recognition technology and contributions to fundamental science. Our project will provide an excellent training ground for graduate students and researchers, with the real possibility of significant commercial benefit to the nation. The deployment of our system in the community will greatly enhance the defence and police forces ability for surveillance and security, and will provide new assistive aids to improve the quality of life and safety for the elderly and disabled.Read moreRead less
Enhanced Multilingual Speaker Recognition through the Incorporation of High-Level Features, Late Fusion and Discriminative Classification Methods. The development of robust multilingual speaker recognition systems will benefit the community through the elimination of fraud incurred by financial institutions and customers by enabling several person authentication applications such as: voice based signatures and document issuance; credit card verification by voice and secure over-the-phone financi ....Enhanced Multilingual Speaker Recognition through the Incorporation of High-Level Features, Late Fusion and Discriminative Classification Methods. The development of robust multilingual speaker recognition systems will benefit the community through the elimination of fraud incurred by financial institutions and customers by enabling several person authentication applications such as: voice based signatures and document issuance; credit card verification by voice and secure over-the-phone financial transactions. The technology will also assist in the protection of the community and safeguard Australia by enabling the implementation of the following: suspect identification using voice print; national security measures for combating terrorism by using voice to locate and track terrorists; preemptive criminal activity counter-measures; surveillance and secure building access by voice.Read moreRead less
Robust speaker recognition with reduced utterance duration and intersession variability. The development of robust and accurate speaker recognition systems will enable secure person authentication in over-the-phone financial transactions and benefit the community through the elimination of identity fraud incurred by customers and financial institutions. The technology will also assist in safeguarding Australia by enabling the implementation of suspect identification using voice and security meas ....Robust speaker recognition with reduced utterance duration and intersession variability. The development of robust and accurate speaker recognition systems will enable secure person authentication in over-the-phone financial transactions and benefit the community through the elimination of identity fraud incurred by customers and financial institutions. The technology will also assist in safeguarding Australia by enabling the implementation of suspect identification using voice and security measures for combating terrorism by using voice to locate and track terrorists. Our research at QUT Speech Research Lab is at the forefront of development in this field and will provide Australia with a technological advantage in the rapidly evolving global market for speaker recognition technology for person authentication applications.Read moreRead less
Robust Automatic Speaker Diarisation of Audio Documents by Exploiting Prior Sources of Information. Speaker Diarisation, the task of determining who spoke when, is a technology fundamental in deriving intelligent information from audio and multimedia resources. The requirement for efficient and accurate Speaker Diarisation systems, portable across different domains is heightened by the explosive growth of audio and multimedia archives online and throughout the world. This research will provide t ....Robust Automatic Speaker Diarisation of Audio Documents by Exploiting Prior Sources of Information. Speaker Diarisation, the task of determining who spoke when, is a technology fundamental in deriving intelligent information from audio and multimedia resources. The requirement for efficient and accurate Speaker Diarisation systems, portable across different domains is heightened by the explosive growth of audio and multimedia archives online and throughout the world. This research will provide the foundation for a commercial service of automatic Speaker Diarisation to be developed, growing Australia's impact on the information and communications technology (ICT) sector. The outcome of this research will also assist in the tracking of terrorist and unlawful activity by enabling effective intelligence gathering from different audio sources.Read moreRead less