Researchers Develop AI Model that is a ‘Shazam’ for Whale Calls

AI-powered whale tracking breakthrough: Researchers have identified the source of a mysterious underwater sound and developed an AI model to track elusive whale species in the Pacific Ocean.

• The mysterious sound, dubbed “biotwang,” was recorded in 2014 near the Mariana Trench and later linked to Bryde’s whales.
• Ann Allen from the National Oceanic and Atmospheric Administration (NOAA) saw an opportunity to use this discovery for tracking migrating Bryde’s whales.

Technological innovation in marine research: Google researchers collaborated with NOAA to develop an AI model capable of identifying vocalizations from eight different whale species.

• The AI model analyzes spectrograms, which convert sound to visual form, to identify whale calls.
• This technology functions similarly to the music-identifying app Shazam, comparing new calls to a trained dataset.
• The AI successfully identifies calls from humpback, blue, fin, North Pacific right, North Atlantic right, minke, killer, and Bryde’s whales.

Ecological insights and implications: The AI-powered analysis has revealed important information about Bryde’s whale behavior and potential challenges faced by the species.

• Biotwangs are most consistently observed among a specific population of Bryde’s whales in the western Pacific.
• The pattern of calls suggests these whales may follow the movements of the transition zone chlorophyll front, an ocean boundary rich in prey.
• This finding highlights the importance of this oceanic feature for whale migration and feeding patterns.

Climate change concerns: The research underscores potential threats to whale populations due to changing ocean conditions.

• Climate change is expected to cause more frequent and extreme El Niño and La Niña events.
• These events may cause the transition zone chlorophyll front to move further north and become more variable.
• As a result, whales may need to travel further and expend more energy to find food, potentially impacting population health.

Advancements in passive acoustic monitoring: The discovery of the biotwang sound and subsequent AI development have significant implications for marine research methods.

• Caroline Casey from the University of California, Santa Cruz, emphasizes the importance of discovering new call types for species that are rarely visible.
• Passive acoustic monitoring allows researchers to detect the presence of these elusive species without direct observation.
• This technique is particularly valuable for studying Bryde’s whales, which are constantly on the move and rarely seen.

Data-driven approach: The research leverages an extensive dataset to improve understanding of whale behavior and distribution.

• Researchers analyzed over 180,000 hours of underwater recordings from NOAA’s network of hydrophones on the Pacific seafloor.
• The AI model’s ability to process this vast amount of data efficiently represents a significant advancement in marine mammal research capabilities.

Interdisciplinary collaboration: The project showcases the power of combining expertise from different fields to solve complex scientific challenges.

• The collaboration between NOAA marine biologists and Google AI researchers demonstrates the potential for tech industry partnerships in environmental research.
• This interdisciplinary approach may serve as a model for future projects addressing similar ecological questions.

Broader implications for marine conservation: The AI-powered whale tracking system has the potential to enhance conservation efforts and inform policy decisions.

• Improved understanding of whale migration patterns and habitat use can help guide marine protected area designations and shipping route regulations.
• The technology could be adapted to study other marine species, contributing to a more comprehensive understanding of ocean ecosystems.
• Real-time monitoring capabilities may enable faster responses to potential threats or changes in whale behavior.