ROYAL MELBOURNE INSTITUTE OF TECHNOLOGY - DSI RESEARCH
I am currently working on a research grant under the supervision of Assoc. Prof. Margaret Lech, which involves conducting research and developing novel DSP algorithms for the Defence Science Institute (DSI) in areas of Speech and Emotion Recognition. My other responsibilities include the analysis and processing of complex speech interactions such as noise removal, voice activity detection, vocal track separation and acoustic features extraction. Expected project completion - March 2016.
The research focuses on the development and evaluation of a software tool for real-time mental and emotional state tracking using speech analysis. Potential applications include training, performance prediction and monitoring, as well as pervasive monitoring for early detection of adverse emotional patterns, which may indicate impending depression or potential anti-social behaviours.
PSYCHOLOGY NETWORK - KRITON SPEECH
I was contracted to develop a speech recognition software (currently known as KRITON) from scratch for automatic knowledge acquisition. The software allows a user to build ontologies by having an interactive dialogue with a computer.
Voice User Interfaces (VUI) allow a user to perform other tasks while s/he is interviewed by the knowledge acquisition system. KRITON Speech takes the VUI concept to the extreme: there are start and stop buttons, everything else is done by speech. The software was designed for users unfamiliar with artificial intelligence (AI) techniques as well as trained knowledge engineers. The first phase of the project was successfully completed and delivered in June 2015. Further developments are currently in progress
DSI Research Group
+ Margaret Lech (Associate Professor)
+ Melissa Stolar (PhD Candidate)
+ Abdel Yussef (Research Assistant)
+ Joachim Diederich (Psychologist)
The video below includes examples of real-time automatic emotion recognition and explains the features of the Mental State Tracker. The speech is read from audio files of the Berlin Emotional Speech database.
ROYAL MELBOURNE INSTITUTE OF TECHNOLOGY - MASTER’S THESIS
Speech acoustic parameters have been reported to provide highly efficient diagnostic cues for depression. Current studies have been focused on characterizing the speech of adult patients but it is known that adult speech differs significantly from adolescent speech, and that the onset of depression is likely to occur during adolescence itself.
My research investigates and explains the differences in speech acoustic parameters characterizing the speech of depressed and non-depressed adolescents, namely F0 (fundamental frequency), jitter, shimmer, log energy, spectral centroid, spectral entropy and glottal pulse duration. Speech data was collected from adolescents aged 14 – 18 and included 68 (49 females and 19 males) diagnosed with major depression and 71 (44 females and 27 males) diagnosed as non-depressed. The speech recordings were made during three different types of family interactions: event-planning, problem-solving and family consensus.
+ Abdel Yussef (RMIT University)
+ Joachim Diederich (Psychology Network)
+ 9th International Conference on Signal Processing & Comm Systems
This project involved the creation of a prototype helmet that monitors environmental conditions in real-time around a miner. The helmet was fitted with various sensors such as pressure, temperature, toxic gas level and light level sensors. Monitoring was done wirelessly using XBEE modules coupled with an Arduino PCB. All the data received were recorded and displayed on a laptop programmed with a MATLAB interface.
+ Abdel Yussef, Catherine Sandoval Rodriguez, He Gong