Researchers devise tool for understanding dog barks, sort of

We've no idea what's up with Europe and its admittedly zany fascination with canine research, but not even two months after a team of Austrians claimed to have trained a litter to use computers, a crew from Hungary is now trumpeting its dog translation software. More specifically, Csaba Molnár (and colleagues) from Eötvös Loránd University have reportedly been able to create an application that can "identify and differentiate the acoustic features of dog barks, and classify them according to different contexts and individual dogs." The gurus began by feeding some 6,000 barks from 14 Hungarian sheepdogs in a half dozen situations to a computer, and during analysis, it was able to correctly classify the barks between 40 and 50-percent of the time. Granted, that's a heck of a lot better than any of us could do, so we suppose we can't yelp about the results too much.

[Via TGDaily, image courtesy of CalgaryAgility]