The Grooved Whale Project

The Grooved Whale Project encapsulates over 25 years of decoding the complex vocalizations of humpback whales   It serves as a unifying framework for a wide range of projects I’ve undertaken, from analyzing whale song through the lens of mathematical patterns, to composing interspecies music to interactive concerts in the wild. My work is a deeply interdisciplinary endeavor that bridges science, art and culture. It draws from bioacoustics, information theory, and artificial intelligence extending even into SETI and METI research as a speculative blueprint for universal language.  

My research is driven by a deceptively simple question: Can whale song be considered music? This inquiry has led me into the evolutionary roots of musical structure, tracing how pattern, repetition, and symmetry, core elements of human composition, also emerge in the vocal traditions of humpback whales. Integrating both deep learning and musical analysis, my work focusses on exploring universal architectures and templates for communication.

As a Whale Song analyst, my work is driven by the need for a unified, data‑driven framework for understanding the full structure and meaning of humpback whale vocalizations. Current approaches often isolate individual dimensions (unit, phrase, theme, song/non-song) without integrating them into a coherent system capable of revealing the deeper architecture of whale communication. My goal is to bridge that gap.

I’m developing a computational framework that draws from music analysis, machine learning, and bioacoustic signal processing to parse whale song beyond its fundamental components. By combining time–frequency decomposition, segmentation algorithms, feature extraction, clustering, and sequence modelling, I’m building a framework and toolkit that can reveal patterns and organizational principles that have been difficult to detect with traditional methods.

I use the same tools that I apply to analyzing humpback whale song (time‑frequency modelling, segmentation, pattern detection, and machine‑learning–based structure analysis) to create hybrid musical works that blend whale and human sound. By breaking whale song into units, phrases, themes, and rhythmic patterns, I can understand its internal logic and then translate those structures into a hybrid musical form. 

This approach lets me compose with whale song "logic" rather than simply layering their recordings into a track or mimicking their sounds, allowing the resulting music to reflect the whales’ own rule systems while introducing human interpretation. The result is a creative process grounded in data, shaped by interspecies patterning, and designed to highlight the shared architecture of communication between humans and humpbacks.