The project
Voice-Based Parkinson’s Disease Detection Using Machine Learning
Rianna explored the intersection of medicine, speech analysis and machine learning. Her project investigated how measurable characteristics of a voice recording can be used as inputs to a classification model.
She brought the work together in Voxora, a web application that allows users to upload or record audio, processes voice features and displays a model prediction. Her report documents the design, implementation, testing and deployment of the prototype.
How the project took shape
Explore voice features
Investigate acoustic characteristics such as pitch, intensity, jitter and shimmer, and how to extract them from speech recordings.
Connect the model
Build a processing pipeline that turns audio into numerical features and passes them to a trained machine learning classifier.
Build a working application
Connect a Vue interface to a Python API, test recording and upload workflows, and deploy the application online.
Project outcomes
- Produced a written project report explaining the question, methods, implementation and future directions.
- Developed Voxora, a web prototype with audio recording, file upload and model-result display.
- Gained practical experience in integrating machine learning, speech processing and web development.
Rianna’s project report
Explore the system design, methodology, implementation and reflections from Rianna’s project.
PDF · 10 pages · July 2026
Download project report (PDF)
