PhD candidate · Researcher · IT educator
James Cook University, Townsville, Australia
I'm a final-year PhD candidate in Information Technology at James Cook University, where I also teach into the IT program. I develop machine learning methods and software that help researchers explore large environmental audio datasets and find sound events for expert review.
My research focuses on lightweight, unsupervised approaches to ecoacoustic segmentation and annotation: finding useful structure in recordings without relying on labelled training data. My broader interests include software development, augmented reality, and practical applications of AI in education.
An open-source tool for the annotation and visualisation of large ecoacoustic datasets. LEAVES combines clustering, interactive visualisations, and expert labelling to make soundscape exploration and annotation more efficient.
Repository · Project website · Paper — Ecological Informatics, 2025
🎥 Watch my LEAVES talk and demonstration at Cornell’s BioacousTalks, hosted by the K. Lisa Yang Center for Conservation Bioacoustics at the Cornell Lab of Ornithology.
Published in Ecological Informatics (2026), this research presents an efficient pipeline for grouping natural soundscape segments using acoustic features, dimensionality reduction, and clustering. The work supports the analysis of biological, environmental, and human-generated sounds.
Repository · Paper — Ecological Informatics, 2026
My current research investigates how changes in acoustic features and relationships between frequency bands can identify candidate sound events in both time and frequency. The aim is to produce boundaries that follow the recording's acoustic structure and help experts locate events worth reviewing.
The output is a set of candidate event regions on a spectrogram; species identification remains a separate annotation or classification step.
I teach across programming, databases, networking, cybersecurity, and cloud computing. I'm interested in practical, progressive learning activities that help students develop and demonstrate their problem-solving skills.
My current education interests include AI-assisted software development, assessment design, and helping students critically evaluate, debug, and explain the code they produce.
| Area | Languages and tools |
|---|---|
| Research and machine learning | Python, scikit-learn, PyTorch, TensorFlow, librosa, Jupyter |
| Web development and visualisation | JavaScript, HTML, CSS, Django, Flask, Dash, Plotly |
| Interactive applications | C#, Unity, Google ARCore |
| Development and research workflow | Git, GitHub, GitLab, VS Code, PyCharm, LaTeX |
Other tools and technologies I've worked with or explored
Java, R, Express.js, Linux, Ubuntu, ARKit, Android Studio, Sublime Text, Electron, TypeScript, AWS, and Docker.
- Email: thomas.napier@jcu.edu.au
- Website: thomasnapier.com
- LinkedIn: linkedin.com/in/thomas-napier
- GitHub: github.com/thomasnapier