[brightcove video="1571336416001" /]
Soon you may be able to control some of your favorite apps using your voice thanks to AT&T.
For decades AT&T researchers have been working on a speech recognition technology called Watson. The software takes speech and turns it into text. While that may seem like a simple-enough process, getting the software to recognize what you're saying regardless of accent, dialect and speech pattern and transcribing those words quickly and accurately is actually a complicated proposition.
Watson is already used in a number of apps and in January AT&T announced it would be partnering with Panasonic and QNX to bring Watson speech recognition to cars.
Now that the technology is closer to perfection, AT&T has decided to make the API for Watson public, putting the tech in the hands of any developer who may want to add voice-control capabilities to apps.
Starting in June developers will have access to a number of different Watson APIs based on specific categories. The idea behind the categories is that speech recognition works better when it knows what types of words to expect. So, you would use one API for local business search, and another for controlling AT&T's U-verse video programming. AT&T will also offer APIs pertaining to basic web searches, Q&A, voicemail to text, general dictation and text messaging.
AT&T developers are also reportedly working on future APIs based on gaming and social media.