SUNNYVALE, CA—Calling it a major breakthrough in interpreting natural linguistic patterns, technology company Voxil announced the release Thursday of a sophisticated new speech recognition program that factors in users’ mouths always being full. “Our latest voice recognition software gives any device the ability to accurately understand and respond to how the average user actually speaks: through large mouthfuls of half-chewed food,” said Voxil CEO Caleb Roush, adding that the technology’s complex algorithms achieve a greater than 99 percent accuracy rate by filtering out a range of extraneous noises, such as loud mastication, the crunching of snacks between teeth, the wet smacking of lips, and the myriad gulping sounds that typically accompany spoken language. “Plus, our program’s built-in GPS capabilities aid speech identification by automatically determining if the user is at, say, a Quiznos and speaking through a giant mouthful of meatball sub, or whether they’ve just returned to their office desk from a pizza shop and are breathily talking with their mouth wide open because the bite they took is still too hot to chew. And while we still have a few minor issues to iron out when people eat peanut butter, this is nevertheless a major step forward.” Roush added that the company is already at work on an updated version that will be capable of interpreting the inarticulate moans and grunts made by users after they have eaten too much.
More from The Onion