During the past three years, Dr. Parsa’s group has been working on the development of novel methods of objective speech and audio quality estimation. Assessment of speech and audio quality is an important topic in a number of fields including wired and wireless communications, hearing aids and assistive listening devices, and multimedia processing systems. For example, speech quality measurements are employed to quantify the relative performance of speech and audio coders used in telecommunications. Similarly, sound quality is an important factor in users’ acceptance of and satisfaction with hearing aids and assistive listening devices. A good sound quality measure is therefore an extremely important parameter that will allow efficient evaluation of competitive signal processing technologies and will assist in the development of better signal processing schemes. Dr. Parsa’s group has developed novel methods of objective estimation of speech quality which include an adaptive neuro-fuzzy technique which employs a neural network trained using perceptually relevant features and fuzzy logic rules, and a statistical pattern recognition based speech quality estimator where the same perceptually relevant features were used to train Gaussian mixture density hidden Markov models (GMD-HMMs).
Recent advances in digital signal processing hardware and software have facilitated the implementation of complex algorithms in digital hearing aids and assistive listening devices. These include new methods and algorithms for compression amplification, digital noise reduction, beamforming, binaural processing, and feedback cancellation. In collaboration with researchers at the National Centre for Audiology, Dr. Parsa’s group is evaluating the impact of these algorithms on perceived speech and audio quality through instrumental and behavioural measures. The aim of this body of research is to develop a standardized measure for quantifying the speech quality which will allow clinical audiologists to assess the relative benefits of various devices that offer similar, but not identical, signal processing algorithms.
Dr. Parsa has established a multi-disciplinary laboratory for applied signal processing research (LASPR) through funding from the Canada Foundation for Innovation (CFI) and the Ontario Investment Trust (OIT). LASPR houses state-of-the-art DSP hardware and software development kits from Texas Instruments and supports a number of cross-disciplinary research projects such as the analysis of bat vocalization and echo location signals, analysis of sympathetic nerve discharge signals, and the processing of EEG and voice signals..