One year after receiving a grant to study canine vocalizations, The University of Texas at Arlington researcher Kenny Zhu has compiled the world's largest video and audio catalog of canine barks, growls and whimpers—and discovered a set of words and phonemes dogs use to communicate.
Dr. Zhu, a professor in the Computer Science and Engineering Department, is encouraged by progress his team has made in deciphering what dogs are saying to each other and their human companions. And so is the scientific world.
He and his students—Sinong Wang, Hridayesh Lekhak and Tuan Deng—recently won the Association for Computational Linguistics Outstanding Paper Award for their research.
"Our primary goal is to prove dogs have a language—maybe not as sophisticated as humans, but with linguistic features that we can decipher," Zhu said. "We have made a great deal of progress, and although we don't completely understand dog language yet, we have created the largest canine vocalization set and identified a set of words and phonemes in the Chihuahua, Shiba Inu and Siberian husky that shows that there is scientific merit to our research."
The research holds strong commercial promise and could extend to other species, including cats, crows and livestock. Zhu's findings could also transform training of service animals, which, for example, could one day be able to "speak" to their owners, improving their effectiveness. Beyond understanding household pets, the work could also bring applications to agriculture.
"We want to prove that our findings are solid, then find applications for what we have learned," Zhu said. "In the future, we might be able to apply our methods to chickens or cattle and be able to identify when diseases are beginning to spread through the flock or the herd. If we can understand what the animals are communicating, we can take steps earlier to stop the spread, save farmers money and keep illnesses from affecting public health."
Zhu's canine vocalization set contains video and audio files, along with metadata such as the dog's age and sex and the environment where the sound was recorded.
Because many of the recordings contain background noise, Zhu and his team first use sound separation tools to isolate the dog's voice. They then identify which noises come directly from the dog's mouth, cluster similar sounds and analyze them to pinpoint phonemes. Finally, they examine how those phonemes combine into syllables.
Zhu is also studying how genetics, environment and social situations shape canine vocalizations. In a long-term study, he has found that young dogs share many vocal similarities, but as they age, their sounds begin to diverge across breeds—evidence that genetics plays a role in vocal development. He is also investigating how interactions with humans may influence dogs' linguistic development.
About The University of Texas at Arlington (UTA)
Celebrating its 130th anniversary in 2025, The University of Texas at Arlington is a growing public research university in the heart of the thriving Dallas-Fort Worth metroplex. With a student body of over 42,700, UTA is the second-largest institution in the University of Texas System, offering more than 180 undergraduate and graduate degree programs. Recognized as a Carnegie R-1 university, UTA stands among the nation's top 5% of institutions for research activity. UTA and its 280,000 alumni generate an annual economic impact of $28.8 billion for the state. The University has received the Innovation and Economic Prosperity designation from the Association of Public and Land Grant Universities and has earned recognition for its focus on student access and success, considered key drivers to economic growth and social progress for North Texas and beyond.
— Jeremy Agor, College of Engineering
