Dave Moffat

Dave Moffat

AI Data Science Lead

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

</p>

Posts

Word2EQ: Word Embeddings for Automatic EQ Mixing

less than 1 minute read

Published:

This is a recent paper submitted to the Journal of the Audio Engineering Society. In this paper, we take word embeddings, and map them directly onto EQ parameters, using a Fully-Connected Neural Network. We show that a neural network can learn equaliser settings for completely unknown words, which produce EQ results that are both intutive, and perceptually sound plausable. Further subjective evaluations are required to validate these results, but in principal, the idea of mapping semantic word descriptors directly onto any audio effect parameters. This approach could be developed in the future, rolled out to a number of different semantic approaches to create a suite of semantically driven audio effects.

</p>

YOHO: You Only Hear Once

less than 1 minute read

Published:

The recent work of Satvik Venkatesh, on the YOHO paper. In this recently published paper, we present a neural network approach for audio detection. In this paper transition points, or sonic objects, are identifed directly through the neural network design, rather than the traditional approach of block based processing of audio and performing classification per block. The traditional approach quantizes the classification of the signal, and relies on accurate classification of every time step, which can be problematic in noisy environments. In this approach, the prediction of the model is a regression, of the transition points exactly, which means the model is much less likely to oscillate, and the predictions are generally considered more robust. A rigorous review of this approach, in noise environments, was presented in a paper at NeurIPS. The full paper is available here.

</p>

chapter

conference

Automatic Subgrouping of Multitrack Audio

Published:

Use Google Scholar for full citation

David Ronan, David Moffat, Hatice Gunes, Joshua Reiss, "Automatic Subgrouping of Multitrack Audio." In the proceedings of Proceedings of the 18th International Conference on Digital Audio Effects (DAFx-15), 2015.

Modal Synthesis of Weapon Sounds

Published:

Use Google Scholar for full citation

Lucas Mengual, David Moffat, Joshua Reiss, "Modal Synthesis of Weapon Sounds." In the proceedings of Proceedings of the Audio Engineering Society Conference: 61st International Conference: Audio for Games, 2016.

Real-Time Physical Model for an Aeolian Harp

Published:

Use Google Scholar for full citation

Rod Selfridge, David Moffat, Joshua Reiss, Eldad Avital, "Real-Time Physical Model for an Aeolian Harp." In the proceedings of Proceedings of the 24th International Congress on Sound and Vibration, 2017.

Unsupervised Taxonomy of Sound Effects

Published:

Use Google Scholar for full citation

David Moffat, David Ronan, Joshusa Reiss, "Unsupervised Taxonomy of Sound Effects." In the proceedings of Proc. 20th International Conference on Digital Audio Effects (DAFx-17), 2017.

Artificially synthesising data for audio classification and segmentation to improve speech and music detection in radio broadcast

Published:

Use Google Scholar for full citation

Satvik Venkatesh, David Moffat, Alexis Kirke, Gözel Shakeri, Stephen Brewster, Jörg Fachner, Helen Odell-Miller, Alex Street, Nicolas Farina, Sube Banerjee, Eduardo Miranda, "Artificially synthesising data for audio classification and segmentation to improve speech and music detection in radio broadcast." In the proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021.

RadioMe: Supporting Individuals with Dementia in Their Own Home… and Beyond

Published:

Di Campli San Vito, P., Brewster, S., Venkatesh, S., Miranda, E., Kirke, A., Moffat, D., Banerjee, S., Street, A., Fachner, J. and Odell-Miller "RadioMe: Supporting Individuals with Dementia in Their Own Home... and Beyond?" In the proceedings of 2022 CHI Conference on Human Factors in Computing Systems (CHI 22) Workshop 32, New Orleans, LA, USA, 30 Apr 2022.

journal

TLS2trees: a scalable tree segmentation pipeline for TLS data Permalink

Published:

Phil Wilkes, Mathias Disney, John Armston, Harm Bartholomeus, Lisa Bentley, Benjamin Brede, Andrew Burt, Kim Calders, Cecilia Chavana-Bryant, Daniel Clewley, Laura Duncanson, Brieanne Forbes, Sean Krisanski, Yadvinder Malhi, David Moffat, Niall Origo, Alexander Shenkin, Wanxin Yang, "TLS2trees: a scalable tree segmentation pipeline for TLS data." Methods in Ecology and Evolution. Wiley. October 2023. https://doi.org/10.1111/2041-210X.14233
Download here

Drone imagery and deep learning for mapping the density of wild Pacific oysters to manage their expansion into protected areas Permalink

Published:

Access paper here

Aser Mata, David Moffat, Sílvia Almeida, Marko Radeta, William Jay, Nigel Mortimer, Katie Awty-Carroll, Oliver R. Thomas, Vanda Brotas, Steve Groom "Drone imagery and deep learning for mapping the density of wild Pacific oysters to manage their expansion into protected areas." Ecological Informatics. 102708, July 2024 https://doi.org/10.1016/j.ecoinf.2024.102708
Download here

Sensing human health from space: An assessment of applications and big data platforms

Published:

Access paper here

D. Sengupta, F. G. Bandao, S. Sathyendranath, G. Kulk, A. Conte, C. Ippolitti, L. Candeloro, D. Moffat, W. Wint, M. Maranesi, R. Scarano, J. Vitorino, G. Brandt and T. M Harish "Sensing human health from space: An assessment of applications and big data platforms." Remote Sensing Applications: Society and Environment 2025 https://doi.org/10.1016/j.rsase.2025.101701
Download here

ForestScan: a unique multiscale dataset of tropical forest structure across 3 continents including terrestrial, UAV and airborne LiDAR and in-situ forest census data Permalink

Published:

Chavana-Bryant, C., Wilkes, P., Yang, W., Burt, A., Vines, P., Bennett, A. C., Pickavance, G. C., Cooper, D. L. M., Lewis, S. L., Phillips, O. L., Brede, B., Lau, A., Herold, M., McNicol, I. M., Mitchard, E. T. A., Coomes, D. A., Jackson, T. D., Makaga, L., Milamizokou Napo, H. O., Ngomanda, A., Ntie, S., Medjibe, V., Dimbonda, P., Soenens, L., Daelemans, V., Proux, L., Nilus, R., Labrière, N., Jeffery, K., Burslem, D. F. R. P., Clewley, D., Moffat, D., Qie, L., Bartholomeus, H., Vincent, G., Barbier, N., Derroire, G., Abernethy, K., Scipal, K., and Disney, M. "ForestScan: a unique multiscale dataset of tropical forest structure across 3 continents including terrestrial, UAV and airborne LiDAR and in-situ forest census data." Earth System Science Data, 18, 1243–1274 February 2026. https://doi.org/10.5194/essd-18-1243-2026
Download here

Modelling primary production: multitude of theories, or multitude of languages? Permalink

Published:

Skákala, J., Sathyendranath, S., Artioli, Y., Banerjee, D. S., Bouman, H., Brewin, R. J. W., Butenschön, M., Ciavatta, S., Dutkiewicz, S., Fidai, Y., Ford, D., George, G., Guihou, K., Jönsson, B., Bačeković Koloper, M., Kovač, Ž., Krishnakumary, L., Kulk, G., Laufkötter, C., Lessin, G., Mattern, J. P., Melet, A., Mignot, A., Moffat, D., Monteiro, F., Rodriguez Bennadji, M., Rousseaux, C., Swaminathan, R., Ulloa, O., and Tjiputra, J. "Modelling primary production: multitude of theories, or multitude of languages? ", Ocean Science, 3.22; EGUsphere, 2026. 1-39 https://doi.org/10.5194/os-22-1457-2026
Download here

portfolio

talks

Deep-learning detection of Harmful Algal Blooms

Published:

The detrimental effects of harmful algal blooms (HABs) on the marine ecosystem, human health, and shellfish and aquaculture industry are well known. Anthropogenic activities have led to an increase in frequency, extent and magnitude of HAB activity. As a result, the detection, monitoring and forecasting of HABs are key to agencies and marine managers, allowing them to implement prevention and remediation strategies.

</p>

Challenges and Solutions for using Machine Learning with Earth Observation Data

Published:

The application of artificial intelligence (AI) and machine learning (ML) is rapidly growing across Earth Observation (EO). State-of-the-art ML techniques can be used to analyse and exploit vast quantities of data, to produce greater insight into data, and enable better understanding of environmental issues. Over the past year, NEODAAS have been utilising their MAGEO GPU cluster to work with users on a range of projects including harmful algal bloom detection, tree monitoring, monitoring global mangroves, ocean oil-spill detection, road vehicle and ship exhaust tracking and underwater image maerl detection. This range of projects has provided opportunities and insights into common challenges in ML with EO data, and how best to overcome them.

</p>

Deep-learning detection of harmful algal blooms

Published:

The detrimental effects of harmful algal blooms (HABs) on the marine ecosystem, human health, and shellfish and aquaculture industry are well known. Anthropogenic activities have led to an increase in frequency, extent and magnitude of HAB activity. As a result, the detection, monitoring and forecasting of HABs are key to agencies and marine managers, allowing them to implement prevention and remediation strategies. However, HAB events are relatively rare events, that can be challenging to detect. HAB detection with satellite image data improves the coverage and efficiency of tracking HABs. Existing remote sensing-based methods frequently rely on statistical classification algorithms. While comparison with cell concentration in situ data has identified two issues: reduced accuracy for the detection of certain species, and accuracy dependency with satellite training data availability. This talk will present a deep-learning technique to improve the performance of the existing models for HAB detection from ocean colour. To this end, we developed a Machine Learning (ML) system using a few-shot learning approach for the detection of Phaeocystis and Pseudo-nitzschia HABs across the French-English channel. We assessed the performance of the ML model in comparison to in situ cell abundance data. The ML system showed better performance than the S-3 EUROHAB model, with results for the detection of Phaeocystis blooms being particularly promising.

</p>

Computer Science: Connecting Music and Environment Science

Published:

Connecting Music and Environment Science: from philosophical approaches to fluid mechanics, and how computer science research can cover a range of domains. In this talk we will discuss a range of different computer science domains and approaches, and demonstrate how non-traditional career paths can be both more interesting but also more advantageous than traditional software development houses.

</p>

An Introduction to Machine Learning for Earth Observation

Published:

In this talk, we will provide an overview of Machine Learning (ML) and it’s uses in Earth Observation (EO). This talk will introduce the concepts of AI and ML, with the aim of explaining the differences and the fundamental approaches that can be taken in the field, with some simple examples. We will then go on to discuss a variety of different potential approaches than can be taken, and how ML can be used to overcome some consistent challenges in EO, and where it is not necessarily appropriate to use.

</p>

Panel Discussion: Can ML4EO save the world?

Published:

I was a panel member at the opening of the inaugaral Machine Learning for Earth Observation Remote Sensing with the Environment Intelligence Network workshop at University of Exeter. The panel discussed example of how machine learning can enhance Earth observation science, emerging trend in AI or machine learning that are particularly exciting, the greatest challenges of using machine learning for Earth Observation studies, ethical and societal considerations associated with using AI and machine learning for Earth observation and future research priorities in using AI and machine learning for remote sensing in Earth observation science.

</p>

A Career in Data Science

Published:

An invited guest lecture reflecting on careers in data science and the application of computational and machine-learning methods across research domains.

</p>

Challenges in Machine Learning for Earth Observations

Published:

A keynote examining practical and scientific challenges in applying machine learning to Earth observation, including data characteristics, modelling choices, evaluation and robust use in environmental science.

</p>

Machine Learning Approaches to the Ocean Carbon Cycle

Published:

An invited presentation for faculty and PhD researchers on machine-learning approaches to ocean carbon-cycle research, including Earth-observation products, model parameterisation and primary-production estimation.

</p>

AI Foundation Models for Earth Observation: Opportunities, Challenges and the Road Ahead

Published:

A chaired panel session, including an introductory presentation, examining the opportunities and challenges of AI foundation models for Earth observation. The discussion considered scientific applications, trust and transparency, European capability, compute and data infrastructure, skills and partnerships, and the roadmap needed for foundation models to deliver lasting scientific and societal value.

</p>

teaching

Computing, Audio and Music Technology Teaching, 2019–2020

Programme leadership and teaching, University of Plymouth, 2019

As Programme Leader for the BSc Computing, Audio and Music Technology programme, I taught Audio Engineering, Introduction to Programming, Sound for Interactive Media, Creative Systems Design and Build, BA Music Dissertation, and ResM Computer Music. I also served as a personal tutor.

</p>

Earth Observation Data for AI

CDT training, University of Exeter CDT, 2021

A hybrid doctoral training day on using Earth-observation data in AI research.

</p>

Coding for Earth Observation

Online training course, EUMETSAT, 2021

A co-led short course on coding for Earth observation for an international EUMETSAT audience.

</p>

PORTWIMS Summer School

Summer school leadership, PORTWIMS, 2022

I led a five-day summer school and delivered ML for EO training.

</p>

Earth Observation Training Programme

International training programme, Trevor Platt Science Foundation, 2023

I contributed two talks and led approximately half a day of course activity and project-based supervision.

</p>

AI for Earth Observation

Online training course, To confirm, 2025

A small-group online training course on AI for Earth observation.

</p>

Introduction to AI for EO Data – Theory

Conference workshop, ML4EO 2026, hosted by NEODAAS, 2026

An introduction to preparing EO data for ML, classification, regression, segmentation, workflow design and a Random Forest example.

</p>