US EditionPublished • loading... • Updated
Researchers Claim to Identify Word-Like Patterns in Dog Barks
UTA researchers led by Kenny Zhu have transcribed 50 hours of dog vocalizations into syllables to identify phonemes and word-like patterns using AI models.
- This year, UT Arlington's Kenny Zhu leads a project repurposing human-speech models to decode dog vocalizations, reporting potential phonemes and word-like patterns.
- Researchers say the project could illuminate early cognitive and neural steps toward speech-readiness and vocal communication evolution, while study authors warn of ethical concerns and the 'uncanny valley' effect, the BARKS Lab at Eötvös Loránd University in Hungary reports.
- So far, the researchers have transcribed about 50 hours of barks into syllables and identified possible words like cat, cage and leash, which vary by breed and age, supported by Zhu's catalog.
- Researchers say the work could improve animal welfare by helping humans interpret dogs' needs, with results presented at the Joint International Conference on Computational Linguistics, Language Resources and Evaluation.
- Zhu described the long-term aim as expanding to cattle and cats, with researchers at the University of Michigan and Virginia Tech also applying AI to decode animal vocalizations, aiming for a future pet translator.
Insights by Ground AI
65 Articles
65 Articles
From barks to words: Researchers aim to translate dog sounds with AI
Ever wonder what your dog is trying to say? Well, a University of Texas at Arlington researcher is aiming to turn barks, howls and whimpers of man's best friend into intelligible speech—a kind of Rosetta Stone of woof.
·United Kingdom
Read Full Article
Researchers at UT Arlington working to translate dog sounds with AI
Computer scientist Kenny Zhu says he’s compiled the “world’s largest video and audio catalog of canine vocalizations,” with the hopes of turning all of the barks, woofs, and whimpers your dog makes into intelligible speech using AI.
·Philadelphia, United States
Read Full ArticleCoverage Details
Total News Sources65
Leaning Left7Leaning Right3Center40Last UpdatedBias Distribution80% Center
Bias Distribution
- 80% of the sources are Center
80% Center
14%
C 80%
Factuality
To view factuality data please Upgrade to Premium
