- 79 views
- 136 downloads
Machine Learning on Speech Audio for Extracting Indicators of Psychiatric and Neurodegenerative Conditions
-
- Author / Creator
- Tasnim, Mashrura
-
The objective of this thesis is to develop and validate computational models of vocal expressions to predict the severity of (i) psychiatric disorders, such as depression, anxiety, and stress, and (ii) neurodegenerative diseases, like Alzheimer's Dementia (AD) and mild cognitive impairment (MCI). The fundamental assumption of this work is that the above conditions, and possibly others, impact the individual's ability to produce language, and therefore their vocal expressions are distinguishable from those of healthy individuals.
In the quest to better understand and predict various aspects of these conditions, this thesis explores a comprehensive exploration of vocal expressions. Leveraging a variety of datasets spanning psychiatric disorders like depression and anxiety and neurodegenerative diseases such as AD and MCI, this research aims to decode the intricate nuances embedded within vocal tones.
The methodological approach incorporates sophisticated audio analysis techniques with supervised machine learning models trained on labelled speech samples. Through a meticulously designed pipeline, encompassing noise reduction, feature extraction, and model training, the research endeavours to establish connections between vocal cues and mental and cognitive health conditions. This thesis underscores the potential of audiovisual cues as invaluable markers for advancing our comprehension and prediction of mental health conditions and cognitive competencies, thereby paving the way for more effective diagnostic and intervention strategies.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2024
-
- Type of Item
- Thesis
-
- Degree
- Doctor of Philosophy
-
- License
- This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.