AI Helps Researchers Decipher Dogs' Secret Language
Researchers are using AI models trained on human speech to decipher dogs' secret language. The study shows that AI can identify dogs' emotions, sex, and breed with high accuracy, paving the way for a deeper understanding of animal communication.
Researchers are using AI models, originally trained on human speech, to decipher dogs' secret language. The study, conducted by researchers at the University of Michigan, Mexico's National Institute of Astrophysics, Optics and Electronics, shows promising results and was recently presented at an international conference. These results show that today's AI models could be the key to understanding animal language, at least to some extent.
Using an Advanced AI Model
The study uses the advanced AI speech model Wav2Vec2 to identify dogs' emotions, sex, and breed based on their barks. The researchers used two different datasets for training and compared the results: a model trained from scratch solely on dog barks, and one pretrained on human speech and then fine-tuned on dog barks. The model pretrained on nearly 1,000 hours of human speech performed better. This model was then fine-tuned on a dataset consisting of vocalizations (barks) from 74 dogs: 42 Chihuahuas, 21 French Poodles, and 11 Schnauzers.
Promising Results and Future Research
This AI model, trained on both humans and dogs, could identify a dog's emotion with 62% accuracy, breed with 62% accuracy, sex with 69% accuracy, and identify a specific dog from a group with 50% accuracy. All these results outperformed the AI model trained solely on dog barks, suggesting that sounds and patterns from human speech could potentially serve as a foundation for understanding animals.
The researchers hypothesize that a dog's vocalization is related to its context. Existing evidence suggests that the sounds made by monkeys and prairie dogs can be predicted based on the context of the situation. Some of the emotions the researchers are trying to attribute to dogs in this study are aggressive barks, normal barks, negative whines, and negative growls.
Next Steps for the Research
The researchers plan to test more breeds, emotions, and species to understand the technology's full potential. This is the first time human speech models have been used to decipher animal communication, and it could lay the groundwork for understanding animal language. The study is seen as a promising step toward understanding the meaning of dogs' barks, even though it isn't yet definitive.