Translation Manchester are excited to announce the return of our popular monthly webinar series ‘Gateway to Translation (G2T)’!
The “Gateway to Translation” (G2T) webinar series is a joint initiative between the University of Cambridge (Cambridge Academy of Therapeutic Sciences, CATS) and The University of Manchester (Translation Manchester). The aim of these monthly webinars is to stimulate the translation of research into the clinics by educating academics about trends, technologies and research in the commercial sphere and by facilitating collaborations between academia and the medtech, biotech and pharma industry sectors.
This month's talk will explore how AI can be deployed on resource?constrained microcontrollers, enabling intelligent functions on small, low?power devices. Using audio as a running example, the session will unpack the technical challenges, trade?offs, and practical solutions involved in bringing machine learning models onto embedded systems in real?world applications.
Speaker: Luke Taylor, Founder of DeepGate.ai
Chair: Dr Garreth Prendergast, Lecturer in Audiology/Hearing Sciences, University of Manchester
About this webinar
Artificial intelligence is increasingly moving away from the cloud and onto tiny, low?power devices. This webinar will explore how AI can be deployed on resource?constrained microcontrollers, enabling intelligent functions on small, low?power devices. Using audio as a running example, the session will unpack the technical challenges, trade?offs, and practical solutions involved in bringing machine learning models onto embedded systems in real?world applications.
Speaker
Luke Taylor is the founder of DeepGate, a company building ultra-efficient machine learning tooling for microcontrollers and embedded systems. He completed his DPhil at the University of Oxford, where he conducted computational neuroscience research in the Auditory Neuroscience Group with a focus on spiking neural network models. Inspired by the energy efficiency of the brain, his work today focuses on making machine learning smaller, faster, and more energy-efficient for real-world edge devices.
DeepGate builds efficient tooling for developing, optimising, and deploying machine learning models on microcontrollers. The company focuses on reducing latency, memory usage, and power consumption for edge AI applications, while simplifying development workflows for developers. Alongside its tooling platform, DeepGate develops licensable embedded AI models, including object detection, wake-word, and audio enhancement systems.
Who should attend?
This webinar is aimed at:
• Academic researchers (PIs, postdocs, and PhD students)
• Engineers and data scientists working on AI, wearables, or embedded systems
• Researchers interested in translational and industry?facing AI technologies
• Anyone curious about deploying AI beyond the cloud and onto real?world devices
