Researchers from the University of Michigan are leveraging artificial intelligence to decipher canine barks, aiming to determine if a dog is playful or angry.

The researchers are also exploring whether AI can accurately identify a dog’s age, gender, and breed based solely on its vocalizations.

“Advances in AI can be used to revolutionize our understanding of animal communication,” the head of the University of Michigan AI Laboratory, Rada Mihalcea, said to Michigan News. “Our research leverages speech processing models initially designed for human speech. This opens up new possibilities for using AI to understand the subtleties of dog barks.”

AI has significantly advanced the comprehension of human speech, enabling technologies like voice-recognition software to distinguish nuances in tone, pitch, and accent. These capabilities have been achieved by training algorithms on vast datasets of human voices. However, a comparable extensive database for animal vocalizations does not exist.

“Animal vocalizations are logistically much harder to solicit and record,” noted Artem Abzaliev, the study’s lead author. To address the challenge, his team investigated whether existing research on human speech could be adapted for animals. They collected a variety of barks, growls, and whimpers from 74 dogs of different breeds, ages, and sexes, in various situations.

The recordings were then analyzed using a machine-learning model originally designed for human speech. Remarkably, the algorithm performed well, achieving an average accuracy of 70 percent across different tests.

“This is the first time that techniques optimized for human speech have been built upon to help with the decoding of animal communication,” Ms. Mihalcea said. “Our results show that the sounds and patterns derived from human speech can serve as a foundation for analyzing and understanding the acoustic patterns of other sounds, such as animal vocalizations.”