Eavesdropping on Talking Fish: Why Scientists are Listening
Fish may seem quiet, but they actually create a surprisingly active soundscape beneath the waves. Scientists are now using artificial intelligence to decipher the complex conversations of coral reef inhabitants.
For many years, underwater acoustic recorders have been used to capture the ambient sounds of coral reefs. This has allowed researchers to indirectly monitor the activity of the reefs and their inhabitants.
Analyzing these recordings, however, has traditionally been a slow and painstaking process. Scientists often have to manually sift through months of audio recordings, painstakingly searching for the specific sounds of a given species.
A new computer system developed by the Woods Hole Oceanographic Institution has been designed to streamline this process. The innovative system utilizes a neural network trained on vast datasets of known fish sounds. This AI-powered system can automatically scan through recordings at a rate 25 times faster than a human listener.
This technology can even identify sounds in real time. If an unknown fish sound is detected, an onboard camera can identify the species making the call, adding it to the databases.
Researchers are now integrating the system into an autonomous underwater vehicle (AUV), nicknamed CUREE. The AUV will cruise over coral reefs, identifying known fish noises while cataloging any new sounds it encounters.
“For the vast majority of species, we haven’t gotten to the point yet where we can say with certainty that a call came from a particular species of fish,” said Seth Mccammon, the lead scientist on the project. “That’s, at least in my mind, the holy grail we’re looking for. By being able to do fish call detection in real time, we can start to build devices that are able to automatically hear a call and then see what fish are nearby.”
A paper detailing the research was recently published in The Journal of the Acoustical Society of America.
