AI for Non-Human Animal Communication

The past few years have seen an unprecedented surge in both the availability of bioacoustic data and the sophistication of AI/machine learning models. This convergence presents a unique window of opportunity to revolutionize our understanding of animal communication and biodiversity. However, achieving this requires a conscious effort to integrate the disciplines of AI/Machine Learning and Ethology.

This workshop will explore the intersection of artificial intelligence (AI) and bioacoustics, aiming to address challenges in processing complex bioacoustic data and interpreting animal signals in order to advance our understanding of non-human animal communication. Join us for a poster session, keynote talks and a panel discussion as we explore key opportunities to use AI to decipher animal languages and thus deepen our understanding of the natural world.

Workshop Schedule

Time	Event
09:00 - 09:45	Coffee Provided
09:00 - 09:15	Opening Remarks
09:15 - 10:15	Plenary Talk by Laela Sayigh
10:15 - 11:15	Lightning Talks: Part 1
11:15 - 11:30	Break
11:30 - 12:30	Plenary Talk by Oisin Mac Aodha
12:30 - 14:00	Poster Session & Lunch
14:00 - 15:00	Lightning Talks: Part 2
15:00 - 15:15	Break
15:15 - 16:15	Final Plenary Talk by Julie Elie
16:15 - 17:00	Panel
17:00	Conclude

Keynote Speakers

Laela Sayigh

Woods Hole Oceanographic Institution (WHOI)

Dr. Sayigh is a leading expert in cetacean behavior and communication, with over 5,000 citations of her work. She holds a Ph.D. from the MIT/WHOI Joint Program and has dedicated her career to understanding the social behavior and acoustic communication of whales and dolphins. She has been involved with a long-term study of bottlenose dolphins in waters near Sarasota, Florida, for many years, where her work has focused on individually distinctive signature whistles and other aspects of dolphin communication.

Recently, this research has involved playback experiments to free-swimming dolphins, filmed with drones. Given the challenges of studying species that spend most of their lives underwater, she is involved in research that utilizes new technologies, such as non-invasive tags, to study cetacean communication systems.

Oisin Mac Aodha

University of Edinburgh

Oisin Mac Aodha is a Reader (aka Associate Professor) in Machine Learning in the School of Informatics at the University of Edinburgh (UoE). He was a Turing Fellow from 2021 to 2025, currently is an ELLIS Scholar, and a founder of the Turing interest group on biodiversity monitoring and forecasting. His current research interests are in the areas of computer vision and machine learning, with a specific emphasis on 3D understanding, human-in-the-loop methods, and AI for conservation and biodiversity monitoring.

From 2016-2019 he was a postdoc in Prof. Pietro Perona's Computational Vision Lab at Caltech working with the Visipedia team. Previous to Caltech, he spent three years (2013-2016) as a postdoc in the Department of Computer Science at University College London (UCL) with Prof. Gabriel Brostow and Prof. Kate Jones. There he worked on interactive machine learning, where the goal was to design algorithms to enable non-programming scientists to semi-automatically explore events of interest in vast quantities of audio and visual data.

He did both his MSc (with Dr. Simon Prince) and PhD (with Prof. Gabriel Brostow) at UCL and has an undergraduate degree in electronic engineering from the University of Galway in Ireland. Before his PhD, he was a research assistant for one year in Prof. Marc Pollefeys' group at ETH Zurich.

Julie Elie

University of California, Berkeley

Dr. Elie is a computational neuroethologist specializing in the neural mechanisms underlying vocal communication in animals. With over 1,400 citations, her work bridges ethology, bioacoustics, neurobiology, and computational neuroscience. She has made fundamental contributions to understanding how animals produce and perceive entire vocal repertoires, particularly in zebra finches and Egyptian fruit bats – both vocal learning species.

Her research combines rigorous behavioral experiments, neural recordings (including wireless recordings from freely behaving animals), and sophisticated computational analyses to decode how the brain represents and processes communication signals. Her work addresses core questions about individual recognition, the relationship between acoustic structure and meaning, and how neural circuits enable flexible vocal production and categorical perception across diverse call types.

Lightning Talks

Part 1

Pratyusha Sharma

WhaleLM: Finding Structure and Information in Sperm Whale Vocalizations and Behavior with Machine Learning
Adam Chareyre & Haodong Zhang

Supervised vs Self-Supervised Representation Learning for Fin Whale Vocalization Detection
Julia Tan

Scaling Frog Monitoring with FrogID: A Robust Classification Pipeline for Citizen Science Using Bioacoustic Foundation Models
Felipe Parodi

PrimateFace: A Machine Learning Resource for Automated Face Analysis in Human and Non-human Primates

Part 2

Luigi Assom

Associative Syntax and Maximal Repetitions reveal context-dependent complexity in fruit bat communication
Umesh Singla

Latent state dynamics in female Drosophila during social interactions
Donato Crisostomi

Model Merging Improves Zero-Shot Generalization in Bioacoustic Foundation Models
Alef Iury Siqueira Ferreira

Beyond Species-Based Assessments: Automatic Acoustic Recognition of Pollinator Function

Accepted Papers

A Functionalist Framework for Machine Learning in Animal Communication – Jack Terwilliger, Stephan Paul Kaufhold, Federico Rossano
A Practical Audio Preprocessing Approach for Multi-Species Sound Classification – Juan J. Diaz, Santiago J. Vargas, Isabella Bermón, Maria J. Guerrero
A Whisker of Truth: A Multimodal Interdisciplinary Machine Learning Approach to Vocal, Visual, and Tactile Signals in the Domestic Cat – Astrid van Toor, Elin N Hirsch, Susanne Schötz
AI Decodes Chimpanzee Vocabulary Items – Antoine Valet, Quentin Bacquelé, Cedric Girard-Buttoz, Roman M. Wittig, Catherine Crockford
AI for Animal Pain Assessment: a New Challenge for Bioacoustics – George Martvel, Annika Bremhorst, Mayara Travalini de Lima, Stelio Pacca Loureiro Luna, Anna Zamansky
An Expert-in-the-Loop Toolbox for Explainable AI in Animal Communication – Álvaro Vega-Hidalgo, Artem Abzaliev, Nicole Guisneuf, Thore Bergman, Rada Mihalcea
Associative Syntax and Maximal Repetitions reveal context-dependent complexity in fruit bat communication – Luigi Assom
AudioSAM: Automated Annotation of Bioacoustic Soundscapes in the Wild – Zachary Baker, Reece S Shuttleworth, Daniela Rus, Antonio Torralba, Jacob Andreas, Pratyusha Sharma
Beyond Species-Based Assessments: Automatic Acoustic Recognition of Pollinator Function – Alef Iury Siqueira Ferreira, José Mesquita-Neto, Nádia Félix Felipe da Silva, Thierson Couto Rosa
BirdAVES in the wild: individual recognition as a step toward zebra finch communication networks – Aneesh Chauhan, Hugo Loning, Evelien ter Avest, Bhavya Botta
CHAT Jr: Wearable Bioacoustics System For Two-Way Communication Research Between Humans and Cetaceans – Charles Ramey, Thad Starner, Mark Hamilton, Denise L Herzing
Decoding gray wolf (Canis lupus) communication at scale in Yellowstone National Park – Amir Dehbi, Daniel Stahler, Matthew C. Metz, Kira A Cassidy, Jeffrey T Reed, Jeremy SunderRaj, Taylor L. Rabe, Brett Melbourne, Joanna E. Lambert
Dolph2Vec: Self-Supervised Representations of Dolphin Vocalizations – Chiara Semenzin, Faadil Mustun, Roberto Dessi, Pierre Orhan, Gonzalo G. de Polavieja, Yair Lakretz, German Sumbre
Fear-Driven Collective Topology: Comparing Smart-Boid Vietoris–Rips Graphs to Animal Communication Networks via Persistent Features – Guilherme S. Y. Giardini, Carlo R. da Cunha
Fighting Poaching Through Targeted Deep Learning and Sensor Integration – Naveen Dhar, Irina Tolkova
From Pretrain to Primate: Decoding Chimp Nocturnes for Wellbeing – Lara Narbona Sabaté
HAC-Net: Learning Natural Units from Acoustic Change – Quentin Bacquelé, Frederic Theunissen, Mathevon
Hierarchical Representation Learning of Dog Behavior via Single-View 3D Pose Estimation – Nanako Miyai, Takatomi Kubo, Maaya Saito, Kazunori Ohno, Takefumi Kikusui, Miho Nagasawa, Kazushi Ikeda
Keypoint Annotation for Electrocommunication Source Separation with PIKAChU and RAIChU – Kaden Zheng, Sonja Johnson-Yu, Satpreet Harcharan Singh, Denis Turcu, Federico Pedraja, Pratyusha Sharma, Naomi Saphra, Nathaniel Sawtell, Kanaka Rajan
Latent state dynamics in female Drosophila during social interactions – Umesh Singla, Albert Lin, Jonathan W. Pillow, Mala Murthy
Marine Mammals Recognition: a Multi-Modal Framework for Bioacoustic Monitoring – Amine Razig, Soulaymani Youssef, Loubna Benabbou, Pierre Cauchy
Model Merging Improves Zero-Shot Generalization in Bioacoustic Foundation Models – Davide Marincione, Donato Crisostomi, Roberto Dessi, Emanuele Rodolà, Emanuele Rossi
Pawgaze: A Benchmark for Fine-Grained Multimodal Analysis of Canine Behavior – Khush Attarde, Divye Bhutani
Perch 2.0 transfers 'whale' to underwater tasks – Andrea Burns, Lauren Harrell, Bart van Merriënboer, Vincent Dumoulin, Jenny Hamer, Tom Denton
PrimateFace: A Machine Learning Resource for Automated Face Analysis in Human and Non-human Primates – Felipe Parodi
Proposal: Deciphering Electrocommunication with MARL and Unsupervised Machine Translation – Satpreet Harcharan Singh, Sonja Johnson-Yu, Zhouyang Lu, Aaron Walsman, Federico Pedraja, Denis Turcu, Pratyusha Sharma, Naomi Saphra, Nathaniel Sawtell, Kanaka Rajan
QuadForecaster: Diffusion-Based Quadruped Pose Prediction for Animal Communication Analysis – Ian Noronha, Aneeq Chowdhury, Saru Bharti, Upinder Kaur
Redefining Intelligence: From Anthropocentric to Relational AI-Species Communication – Kerri Lake
Retrieval-Augmented Bioacoustics: Evidence-Guided Generation for Animal Communication – Bodhisatta Maiti
Robust detection of overlapping bioacoustic sound events – Louis Mahon, Benjamin Hoffman, Logan S. James, Maddie Cusimano, Masato Hagiwara, Sarah C. Woolley, Felix Effenberger, Sara Keen, Jen-Yu Liu, Olivier Pietquin
Scaling Frog Monitoring with FrogID: A Robust Classification Pipeline for Citizen Science Using Bioacoustic Foundation Models – Julia Tan, Jodi J. L. Rowley
Self-Supervised vs Supervised Representation Learning for Fin Whale Vocalization Detection – Adam Chareyre, Haodong Zhang, Shuwen Ge, Randall Balestriero, Hervé Glotin, Paris Sebastien
SongMAE: Fine-Grained Syllable Discovery in Birdsong Using Asymmetric Patches – George Vengrovski, Timothy J Gardner
Soundscape Connectomes: Unsupervised Graph-Based Approach for Soundscape Mapping – Maria J. Guerrero, Aref Einizade, Jhony H. Giraldo, Victor M. Martinez-Arias, Claudia Isaza, Cesar A Uribe
The Double Contingency Problem: AI Recursion and the Limits of Interspecies Understanding – Graham L. Bishop
The responsible media representation of artificial intelligence for non-human animal communication – Caatje Kluskens
Toward Multimodal Models of Animal Communication: Insights from New York City Rats – Ralph E Peterson, Dmitry Batenkov, Ahmed El Hady, Emily L Mackevicius
Toward Visualization of EEG Signatures for Anomaly Detection in Dogs Using CEBRA – Natsuna Ishida, Takatomi Kubo, Hitomi Koike, Yuki Maruno, Maaya Saito, Kazunori Ohno, Toshitaka Yamakawa, Takefumi Kikusui, Miho Nagasawa, Yuichi Sakumura
Towards Leveraging Sequential Structure in Animal Vocalizations – Eklavya Sarkar, Mathew Magimai Doss
Understanding Electro-communication and Electro-sensing in Weakly Electric Fish using Multi-Agent Deep Reinforcement Learning – Satpreet Harcharan Singh, Sonja Johnson-Yu, Zhouyang Lu, Aaron Walsman, Federico Pedraja, Denis Turcu, Pratyusha Sharma, Naomi Saphra, Nathaniel Sawtell, Kanaka Rajan
WhaleLM: Finding Structure and Information in Sperm Whale Vocalizations and Behavior with Machine Learning – Pratyusha Sharma, Shane Gero, Daniela Rus, Antonio Torralba, Jacob Andreas

Call for Papers

View: Submission Details, Tracks, and Evaluation Criteria

We invite short papers and proposals papers related to using AI for non-human animal communication. The following are some relevant areas for consideration, but proposals are not limited to these topics.

Data
- Benchmarking and evaluation datasets
- Multimodal data
- Study designs for animal communications (passive/long duration versus cleaned specific examples versus experiments)
- Repositories and data sources
AI advancements
- Bioacoustics
- Multimodal learning
- New model architectures
- Transfer learning
- Large-language models
- Unsupervised, self-supervised, or supervised models
Applications and case studies
- Ecological impact and conservation monitoring
- Specific species studies
- Generalizability across species
Communication and cognition
- Linguistics
- Information encoding
- Experimental design
- Ethics

Important Dates

September 5, 2025 23:59 (AoE)

Paper Submission Deadline

~~September 22, 2025~~
October 1, 2025

Notification of Acceptance/Rejection

November 15, 2025

Camera-ready Materials Due

Submission Portal

Please use NeurIPS 2025 LaTeX style file for submissions.
Submissions must be anonymized.
The review process will be double-blind.

Submission Deadline Passed

The submission deadline was September 5, 2025. Submissions are no longer being accepted.

Call for Reviewers

We also welcome applications to review, please express your interest to aiforanimalcomms@earthspecies.org.

Submission Tracks

We have two tracks for submissions of manuscripts: short papers and proposals. Submissions to this workshop are considered non-archival and thus may be submitted elsewhere in the future, but should represent novel work not previously published.

Papers Track

We welcome submissions of brief research papers that summarize the context, methods, and results of a study or set of experiments. Please limit your submission to 4 pages excluding references and supplementary materials. Reviewers will not be required to read or review the supplementary content and will evaluate the papers based on the main content. Papers should have the following components:

Introduction/motivation - what is the study objective and why does it matter?
Methodology - describe the methods and/or models used
Results - summarize key findings
Conclusions

Proposals Track

The proposals track is intended for articulating new ideas for research or forward-looking advancements that do not have finalized results, or for proposing how concepts from interdisciplinary or domain research would apply in AI development or applications in the domain of animal communication.

Evaluation Criteria

Impact and significance of the research — Is the question well-motivated by prior research? Can the methods and findings be generalized to other species or studies? Does the research resolve a previously unresolved challenge in the use of AI?
Collaboration — does the research involve collaboration with domain experts and scientists outside of the field of AI?
Appropriateness of the methodology
Clarity of communication
Ethical considerations

FAQs

Can I submit work I've published previously?

Submissions should be novel work not previously published within machine learning conferences - however, new findings that build on prior work are welcome. We may consider new presentations of work previously published in applied scientific journals on a case-by-case basis as long as the findings were not previously published.

What happens if my submission is accepted?

We will be hosting a poster session and potentially a short series of talks on successful papers (the schedule for the day is still TBC). Please note that talks will take place in person; we are not able to support virtual presentations.