Music lookup services like Shazam have been helping listeners get songs out of their heads and into their music players for years, but now researchers are realizing that there's even more scientific value to be had from the principles behind these apps. Using the very algorithm employed by many music lookup services, researchers have found that it's possible to distinguish between Dolphins' "signature whistles" — the sounds that they use to effectively address other dolphins by name — in a simple and efficient way.
"Vocalizations are highly varied."
Researchers from the National Institute for Mathematical and Biological Synthesis are reporting the findings today in a paper published in PLOS One. The researchers examined 400 whistles from 20 different dolphins with the algorithm that's behind many music identification services, and they found that it was actually able to distinguish between dolphins' calls more quickly than current methods. Dolphins' whistles have traditionally been analyzed using a spectrograph, but spectrographs look at a range of information that isn't really necessary, making the process take longer than it needs to.
The new method is also said to do a good job at identifying the variations between different dolphins calls. "Vocalizations are highly varied and presumably serve varied functions," Arik Kershenbaum, lead author of the paper, says in a statement. By examining the differences in dolphins' signature whistles, researchers may be able to better study how their whistles are affected by others around them. "Determining what aspects of the vocalizations hold information is crucial to be able to classify them and to be able to understand their meaning."