Intel Relevant Products/Services sees "perceptual computing" as the next big wave for personal computing. That vision, articulated by Executive Vice President Dadi Perlmutter in a keynote speech on Tuesday at the Intel Developer Forum in San Francisco, echoes similar futuristic landscapes recently envisioned by IBM and Intel competitor Advanced Micro Devices.

In Perlmutter's view, we're in the process of leaving behind mice and keyboards. Touchscreens, which have gotten new life following the success of touch-sensitive tablets, and new attention with the coming full release of Microsoft Relevant Products/Services's Windows 8 operating system, are "just the beginning," he said. In Intel's view, this new perceptual-based interactivity will instead be built around voice commands, facial recognition, eye-tracking and gestural controls.

'More Natural and Intuitive'

"People love the way" they can interact with machines using voice and gesture, Perlmutter said, adding that the new perceptual computing "will be more natural and intuitive."

At the conference, Perlmutter demonstrated new, more powerful voice-recognition technology from Nuance. Intel has also been collaborating with a company called SoftKinectic, whose aim is to deploy Relevant Products/Services gestural recognition on PCs that can distinguish between each of one's 10 fingers. Perlmutter demonstrated a catapult game where fingers could be held and gestured in front of the laptop to hold a virtual Relevant Products/Services crystal ball.

Intel has also been promoting the coming of the Internet of Things, where all kinds of non-PC, non-smartphone devices have built-in intelligence Relevant Products/Services and Net connectivity -- resulting in the possibility of a natural interface that would encompass virtually all of one's environment, not just the desktop, phones, a laptop or a TV.

Interestingly, a similar view of the Next Big Thing in personal computing was sketched at the end of last month by AMD's chief technology officer, Mark Papermaster.

Papermaster shared his vision for what he called Surround Computing at the Hot Chips semiconductor design conference, held in late August in Cupertino, Calif. He said the industry's last 10 to 20 years have been used to develop computing capability to simulate reality, and the next decade or two will be focused on providing the capabilities to interpret content and contexts in order to provide better experiences.

'Contextual Insight'

The Surround Computing Era, he said, is multi-platform, ranging from eyeglasses to room-sized computing devices. It's also fluid, with realistic output and natural human input, and it's intelligent, anticipating human needs.

This kind of computing, he said, "imagines a world without keyboards or mice, where natural user interfaces based on voice and facial recognition redefine the PC experience," and where devices deliver "contextual insight and value," even as they "disappear seamlessly into the background."

In his view, natural user interfaces will go beyond Microsoft's gestural and facial recognition controller Kinect, and beyond Siri's intelligent voice agent, which, by themselves, already fill out many sci-fi visions. Papermaster is intending to build systems that also employ fingerprints and tactile information, and that can interpret facial expressions and speech better than current technology.

He told The Wall Street Journal that "it's going to be a tsunami" driven by "this inexorable growth of all these sending devices around us."

To be truly natural, computing devices will need an intelligence that is approaching that of humans. In short, a Watson, the IBM-built supercomputer that defeated two champion Jeopardy players on live TV, at their own game.

Recently, IBM Vice President Bernie Meyerson said his company was working on turning Watson into a back-end, intelligent voice agent for smartphones and other devices, one that could be available from virtually any connected device. In short, an anywhere version of Apple's Siri, but on steroids.