Voice recognition systems have been around for a few years now. Though the history doesn’t boast of many breathtaking achievements in this field, there have been points where manufacturers could capture the imagination of users with innovation. Siri holds the distinction of being the most prominent voice recognition system till date. The accuracy of these systems are better than ever, but the cause of concern is the slowness of computation. Intel has reportedly found a solution for this problem. Voice recognition system working on any platform is designed to work with servers, where the device sends compressed signals to the server and waits for response.
In an interview with Quartz, Mike Bell, head of wearable at Intel, said that the chip manufacturer is developing a voice recognition system without cloud. The technique is to localize the processing so that the round-trip to the cloud is ruled out. Intel has partnered with an unnamed third party to put the software on the Intel mobile processors. To demonstrate the feasibility, a prototype wearable headset, Jarvis, has been developed with the built in voice recognition software. The name, Jarvis, seems to be derived from the name of artificial intelligence software in the Ironman franchise.
The company claims that Intel’s solution will be more responsive than other ‘cloud-obsessed’ solutions in the market. Talking about the usefulness of on-device processing of the voice commands, Bell says, “How annoying is it when you’re in Yosemite and your personal assistant doesn’t work because you can’t get a wireless connection? It’s fine if [voice recognition systems] can’t make a dinner reservation because the phone can’t get to the cloud, but why can’t it get me Google Maps on the phone or turn off the volume?”
It will be interesting to see whether Intel will be able to revolutionize the mobile processors. The PC processor giant wants to pitch in hard with its mobile chip capabilities, as it has missed this wagon quite a few times.
Image Source: Extremetech