The Speech and Behavioural Signal Processing Laboratory is known internationally for its research into automatic emotion and mental state inference from speech and behavioural signals, pronunciation detection and speaker and language identification. 

Our laboratory is equipped with: 

  • A large team of senior and early-career academic staff, postdocs, PhD and honours students 
  • High performance computing capabilities and a large library of algorithms/code, scripts and databases of speech and other signals 
  • Smartphone applications for gathering large amounts of data under realistic conditions (via partners) 
  • A new soundproofed, light-controlled studio facility for recording of speech and behavioural signals under a range of different protocols 
  • Our experts are leading research in: 

    • Voice biometrics and anti-spoofing countermeasures 
    • Automatic inference of emotion and distress from speech 
    • Automatic inference of mental state. Examples include cognitive ability and impairment, and depression from speech 
    • Automatic pronunciation detection 
    • Machine learning 
    • Affective computing 

    We translate our research into:  

    • Monitoring mental state via smartphone 
    • Smart health monitoring and interventions 
    • Automated speech therapy and second language learning 
    • Live analysis of web-based remote video consultation 
    • Joint modelling and recognition of linguistic and paralinguistic speech information (DP’11) 
    • Affective Sensing Technology for the Detection and Monitoring of Depression and Melancholia (DP’13) 
    • Automatic Task Analysis for Wearable Computing (US Army ITC-PAC, ‘15) 
    • Investigating Bayesian Frameworks for Paralinguistic Classification (UNSW Engineering ’16) 
    • Automatic speech-based assessment of mental state via mobile device (LP’16) 
    • Integrating Biologically Inspired Auditory Models into Deep Learning (DP’19) 
    • AusKidTalk: An Australian children’s speech corpus (LE’19) 
    • Speech Recognition Adaptation for Low Research Populations (DP’20) 
    • Developing a paralinguistic plus episodic memory screening tool to detect and track cognitive impairment in the elderly (UNSW Biomed Seed Fund ‘20) 
    • Biologically inspired binaural coupling for selective machine hearing (DP’21) 
    • National University of Singapore 
    • Black Dog Institute 
    • Sonde 
    • University of Canberra 
    • MIT Lincoln Laboratory 
    • Australian National University 
    • QIMR Berghofer 
    • Kids Cancer Centre 
    • USC – Signals Analysis and Interpretation Laboratory (SAIL)

Our people

View Profile
View Profile