UVic Researchers Harness AI to Decode Unique Fish Sounds

Biologists from the University of Victoria have made a groundbreaking discovery regarding the communication of fish. Using advanced artificial intelligence, they identified distinct sounds made by various fish species, demonstrating that even closely related species produce unique acoustic signals. This research opens new avenues for understanding marine life through sound.

Using passive acoustics, the researchers focused on eight fish species native to Vancouver Island. They created a machine learning model capable of predicting fish species based on sound with an impressive accuracy rate of 88 percent. “We previously knew that many fish made sounds in the wild, but we didn’t know which sounds belonged to which species,” stated Darienne Lancaster, a PhD student and lead researcher, in a recent news release. “Now, just as we use bird song to identify specific bird species, we can listen to fish sounds to identify specific fish species.”

Among the unique sounds identified, the black rockfish produces a long, growling noise reminiscent of a frog’s croak, while the quillback rockfish emits a series of short knocks and grunts. Lancaster noted the importance of these vocalizations in understanding fish behavior. For instance, the quillback rockfish issues rapid grunting sounds when chased, indicating a possible defensive mechanism. Conversely, the copper rockfish repetitively knocks as it hunts prey on the ocean floor.

The innovative approach taken by the researchers involved passive acoustic monitoring. This technique allowed Lancaster to collect underwater audio and video using a sound localization array developed by former UVic PhD student and collaborator Xavier Mouy. By analyzing sound characteristics, the team identified differences in species-specific calls. The AI model utilized a set of 47 different sound features, including duration and frequency, to distinguish the subtle variations between the sounds of different species.

The implications of this research extend beyond local waters. The techniques developed by Lancaster can be adapted by scientists globally, enabling them to decode fish calls in various marine environments. The study received funding from the Natural Sciences and Engineering Research Council of Canada and Fisheries and Oceans Canada, highlighting its significance in advancing marine biology.

As marine ecosystems face increasing pressures from climate change and human activity, understanding fish communication could play a crucial role in conservation efforts. The ability to identify and monitor fish species through their sounds presents a promising tool for researchers aiming to protect marine biodiversity.