Barrett, Liam;
(2024)
Measurement of feedback in voice control and application in predicting and reducing stuttering using machine learning.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Thesis_LiamBarrett_2-0-2.pdf - Accepted Version Download (26MB) | Preview |
Abstract
How the brain uses feedback information during speech to maintain fluency is a complex and unresolved process. Additionally, how alterations to feedback can be utilized for speech fluency enhancement may not be optimal. Hence, this thesis seeks to a.) further knowledge on feedback in the speech-motor system, b.) utilize machine learning methods to identify and predict moments of dysfluency and, c.) optimize fluency-enhancing methods with the use of machine learning. To this end, the thesis includes original, experimental work on altered sensory feedback in people who do and do not stutter. Mixed-linear modeling has revealed how brain stimulation can alter learning of new speech-motor mappings, providing a platform for therapeutic uses for dysfluent populations. Such research has also demonstrated fluency enhancing effects of somatosensory alterations in people who stutter. Adjunct neuroimaging evidence using functional near infrared spectroscopy links these fluency enhancements to specific areas of the cortex, furthering the understanding of speech’s neural mechanisms. Machine learning models - including Logistic Regression, Support Vector Machines and Deep Neural Networks - were trained to identify stuttering from the audio signal. The work conducted on this has had major impacts on how stuttering is approached from a machine learning perspective. In particular, a systematic review alongside experimental testing has improved standards in the field. Together, the PhD has improved our understanding of feedback in voice control in both fluent and dysfluent speakers. Additionally, how speech-motor control is sub-served by the brain is furthered. The PhD has delivered machine learning methods to identify stuttered speech in conjunction with protocols to enhance fluency. From this work, future research will be able to improve stuttering recognition as well as integrate machine learning and fluency enhancement methods to optimize stuttering prostheses
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | Measurement of feedback in voice control and application in predicting and reducing stuttering using machine learning |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| UCL classification: | UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences > The Ear Institute UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences > Div of Psychology and Lang Sciences > Experimental Psychology UCL |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10190015 |
Archive Staff Only
 |
View Item |
