Researchers are well aware that mice and rats are social. They just talk all day long with each other, but what are they talking? Many rodent vocalizations were developed to identify that but failed miserably. You might be confused why we are saying this?
There is yet again another innovation by two scientists from the University of Washington School of Medicine. They came up with a software named “DeepSqueak” which converts an audio signal into image, or sonogram. This is the first software that used deep artificial neural networks in squeak detection.
"DeepSqueak uses biomimetic algorithms that learn to isolate vocalizations by being given labeled examples of vocalizations and noise," said co-author Russell Marx.
Coffey, a postdoctoral fellow in the Neumaier lab said that rodents will be very happy when they are rewarded with something like sugar, or playing with their peers. He observed that when two male mice are together, they make the same calls repeatedly.
When they feel there is a female mouse close by, their vocalizations become convoluted, like as if they are singing a courtship song. It becomes more dramatic when they sense the presence of a female mouse but cannot see them. While it can be suggested that male mice have different songs for different stages of courtship.
John Neumaier, professor of psychiatry and behavioral sciences at the UW School of Medicine, head of the Division of Psychiatric Neurosciences and associate director of the Alcohol and Drug Abuse Institute, says that his only target is to create treatments that evacuate opioids or alcohol. In a way DeepSqueak will surely help the two researchers in doing something nobody ever did--to make ultrasonic vocalizations globally available, cost-effective and easy to use.