05/20/2026
By Danielle Fretwell

The Francis College of Engineering, Department of Biomedical Engineering, invites you to attend a Doctoral Dissertation Proposal defense by Aliya Newby titled: "An Investigation of Pediatric Impaired Speech: Characterizing Objective Speech Features and Data-Driven Speech Diagnostics."

Candidate Name: Aliya Newby
Degree: Doctoral
Defense Date: Friday, May 29, 2026
Time: 11 a.m. - noon
Location: Falmouth 302 Conference Room. Please email the advisor Lara_Thompson@uml.edu if you would like to attend on Zoom.

Committee:

  • Advisor: Lara Thompson, Professor of Biomedical Engineering, UMass Lowell
  • Walfre Franco, Associate Professor and Chair of Biomedical Engineering, UMass Lowell
  • Kavitha Chandra, Professor of Electrical and Computer Engineering, UMass Lowell

Abstract:
In the United States, speech, voice, and language disorders affect approximately 1 in 14 children aged 3-17 years old. The diagnosis of speech and communication disorders in children is particularly important in that, if impairments are left undiagnosed, they could impact communication, academics, social interactions, and overall quality of life as the child develops. Currently, speech diagnostic approaches rely heavily on highly subjective clinical assessments; these assessments can show variability in accuracy or incorrect diagnosis, as well as delays in diagnoses. Hence, there is a need for objective, quantifiable diagnostic methods for children that can offer guidance while informing personalized and effective, speech training strategies.

Our overarching goals are to identify characteristic markers of voice in children with impaired speech by investigating extracted acoustic features and to distinguish between different speech disorders using trained machine learning (ML) models. We will achieve our research goals by characterizing the key features of select pediatric speech disorders (Aim 1), distinguishing the key features of each using ML models (Aim 2), and exploring methodologies used for speech screening and training in children (Aim 3). To support these aims, recorded audio samples from open-source pediatric speech datasets which include phoneme articulation, word repetition tasks and spontaneous speech will be processed to extract a comprehensive set of time-domain, frequency-domain and nonlinear acoustic features. These features will support the characterization of speech patterns in pediatric patients. The extracted features will subsequently be used in conjunction with classification models including Support Vector Machines (SVM), Random Forest (RF), Convolution Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM). Expected outcomes are: 1) characterization of objective, quantitative acoustic markers which can be used toward voice as a biomarker in children with speech impairments, including categorizing features that are most prominent in various speech impairments, as well as 2) diagnostics in pediatric voice via ML models which identify and distinguish impaired speech. This holds potential for diagnostic applications and the capacity to inform effective methodologies for improving speech articulation in children. Overall, this project aims to provide data-driven metrics to enhance diagnostic approaches toward disorders involving pediatric speech.