Scientists have created a computer system that attempts to recognize human emotions such as anger and impatience by analyzing the acoustics of one’s voice. Such a system would have obvious implications for perennially frustrating interactive voice response systems, but could be applied to other areas as well.
If you were following along at Disrupt SF, perhaps you caught Vocre’s impressive demonstration of their near-real-time spoken translation app. As I was watching, I was picturing the gears turning behind the veneer of the app, though: the cloud transcription, translation, and speech APIs, and how there’s a nice big market for this kind of thing. Google knows it, and of course we’ve had speech on Android for a long time. Apple knows it, but took its time to release it in a more consumer-focused package.
Now even defense contractor Raytheon is getting into the game. Their TransTalk app, which has emerged from the soup of defense contracts and government research funds that is DARPA, is specifically designed for deployment in the middle east.
Apple’s iOS 5 for iPhone, iPad and iPod touch is on the way, and now that the iOS5 beta software is available to developers, clues of a partially hidden feature have been uncovered: Nuance speech recognition, rumored to be tightly integrated within iOS 5.
Although the upcoming speech-recognition features are not operational yet in the iOS 5 beta, parts of it are currently visible in the interface. As you can see in the graphic above, 9to5 Mac points out the microphone icon, placed next to the space bar on the iOS keyboard that’s used throughout the operating system. When a user touches that microphone icon, it opens the speech recognition interface that you see on the right of the graphic above.
Follow