Research Assistant/Associate (Fixed Term) Part-time

Research Associate/Assistant in Machine Learning for Health Audio AI

Fixed-term: The funds for this post are available for up to 12 months with an ideal start date of June/July 2026.

We are seeking a Postdoctoral Research Associate in Machine Learning for Audio to join the AURORA project (AUdio models for RespiratOry and cArdiac diagnostics and clinical tRAining), funded by the ERC Proof of Concept programme.

The project aims to develop machine learning models and software tools for analysing human health sounds, such as respiratory and cardiac audio, to support clinical diagnostics and medical training. Building on results from the ERC Advanced Grant EAR, the project will advance foundation models for health audio, integrate large language models for clinician interaction, and explore the use of synthetic audio generation for clinical education and evaluation.

The successful candidate will contribute to the development of robust machine learning models for cardio-respiratory sound analysis, as well as to the creation of tools that can be evaluated with clinicians and medical students.

Key Responsibilities

The Research Associate will:

• Develop and improve machine learning models for cardiac and respiratory audio analysis
• Work on audio foundation models and multimodal AI systems Investigate model robustness and generalisation across diverse clinical audio datasets
• Contribute to the generation of synthetic cardio-respiratory audio data
• Assist in integrating models into user-facing tools and mobile applications
• Collaborate with clinicians to evaluate models in diagnostic and training contexts
• Contribute to open-source software releases and research publications

Requirements

Applicants should have:

• A PhD (or be close to completion) in machine learning, signal processing, computer science, or a related field OR relevant experience
• Strong experience in machine learning for audio, speech, or acoustic signal processing
• Excellent programming skills (e.g., Python, PyTorch/TensorFlow)
• Experience working with deep learning models
• A strong publication record in relevant venues

Desirable experience includes:

• Machine learning for healthcare or biomedical signals
• Audio foundation models or generative models
• Multimodal AI or LLM integration
• Working with large datasets or open-source ML frameworks

Research Environment

The position will be based in the Department of Computer Science and Technology at the University of Cambridge, a leading centre for research in machine learning, digital health, and mobile sensing. The project is led by Professor Cecilia Mascolo, whose group works on machine learning for health and wearable/mobile sensing technologies.

The successful candidate will collaborate with an interdisciplinary team including machine learning researchers, software engineers, and clinical partners, contributing to the development of tools that could transform audio-based diagnostics and medical training.