Click to View

Although it's taken a long time to get here, PDAs and smartphones finally have enough processing power to implement accurate voice control and speech recognition—that doesn't require training.

"Call Nancy" was the voice-command state of the art until quite recently, but today's mobile devices are capable of much more. This boost in speech power should drive the acceptance of handhelds, while making them safer. As drivers can, for instance, make phone calls or pull up directions on a smartphone without taking there eyes off the road and their hands off the steering wheel.

This week, announcements from three software vendors—Fonix, VoiceSignal, and Microsoft—highlight progress on the voice recognition front.

With Fonix's VoiceIn 4.0 SDK, manufacturers and developers can add Windows Mobile and Linux-based automatic voice control and speech recognition to a number of different types of devices, including mobile phones, smartphones, PDAs, electronic dictionaries, PCs, automobiles, medical devices, and alarm systems.

Unlike older systems, Fonix said its technology provides accurate speaker-independent recognition, even in noisy environments, and does not require the user to train the system to recognize his speech. So users can speak naturally, without pausing.

As result, end-users can access information and operate device functions, such as looking up an address and dialing, without pressing buttons, scrolling through menus, looking at screens or typing.

The SDK also supports nine languages (U.S. and U.K. English, Canadian and European French, Castilian and Latin American Spanish, German, Swedish and Japanese) as well as phonetic voice tags and the dynamic creation of recognition grammars. It can also distinguish words outside its programmed vocabulary.

According to Fonix, while it sells a voice dialing application for Pocket PC phones at retail, it plans to release e-mail, Internet and calendar speech recognition software for Windows Mobile, Palm and Symbian devices shortly.

Another company, VoiceSignal, sells a speaker-independent speech recognition solution that is bundled with the recently shipped Sendo X multimedia phone (see image), one of the most anticipated smartphones of the year.

While the VoiceSignal application is limited to simple "Call Nancy" functionality, it does this on a speaker-independent basis. The Sendo X is also the first smartphone to let users launch the application from a wireless Bluetooth headset.

Sendo CEO Hugh Brogan commented "We believe that as mobile phones become increasingly sophisticated, speaker-independent speech recognition will make these converged devices more convenient and fun to use."

Finally, Microsoft's Voice Command software for Windows Mobile Pocket PC handhelds and phones is now available at retail, marking the first time the company is offering standalone Pocket PC software on store shelves.

Voice Command lets users access a number of different Pocket PC functions via voice commands, including contacts, dialing, calendar, and media player—again, without device training.

New to the software package is compatibility with Hewlett-Packard's upcoming iPAQ h6315 smartphone (see HP Launches Wi-Fi Enabled Smartphone, PDAs).