Voice Pilot offers OS choice

By Owen Ferguson

Rolf Rudestam wants his Voice Pilot Pal to be the Java of voice-enabled applications. And, as the president and CEO of Voice Pilot Technologies Inc., he just might make it happen.

"Our Voice Pilot Pal is the first speech engine independent application on the market. What that really means is we view all the speech recognition engines as operating systems, as platforms, much like, Unix or OS/2 or Windows 95, and it runs on any of them," he said from his office in Big Bear Lake, Calif., where the company is headquartered.

"We are in the speech computing area. Our main product line, where we really have our growth, is in speech aware applications," he says. "We do not consider ourselves to be competing with IBM (ViaVoice) or Dragon System (Dragon Dictate) or Microsoft with their new Whisper engine, because we don't make a speech recognition engine."

Rudestam says that making the application independent of the speech engine is a big step because it means a much more versatile product for the user. "This product signals a new direction for the company and for speech recognition," he adds.

"We will work towards providing enhanced value for the user without inhibiting their particular decision about which speech recognition engine they care to use, because each has their own separate advantages. We will ship that product with the Microsoft Whisper engine, just for those people who say 'I'm not ready to spend $200 on a speech recognition engine, but I'd like to try your suite.'"

Although Voice Pilot Pal won't be available until later in the year, the company is currently shipping a number of other voice-based products, including Voice Pilot Continuous. This is a suite of applications, focusing primarily on a PIM, giving the user the ability to manage things like their appointments and calendar all by voice. It also includes the IBM ViaVoice engine.

Rudestam said his company has been successful because it recognizes that real-time speech recognition engines are more useful for application control than for dictation, because the current engines simply aren't robust enough to handle flat-out text dictation.

"The solution for dictation is more the approach that Sony has taken, and that Phillips has taken, where you get a digital recorder and dictate to that, then let the computer transcribe the dictation later, rather than trying to transcribe on the fly, which requires so many resources," he said.

But voice control isn't Voice Pilot's only focus. The company also owns Voice Pilot Hear-Say, a voice-based e-mail program add-on. The software allows users to record audio greetings, compress them, and attach them, along with playback drivers, to e-mail messages in any e-mail program. The compression ratios available are quite powerful, ranging from 16-to-one to 125-to-one. On the highest compression rating, a one-minute .wav file, which would weigh in at a hefty 1.3MB uncompressed is a mere 10KB.

The company also recently announced Voice Pilot Hear-Look, a further extension of the Hear-Say e-mail enhancement product line that includes the ability to compress and attach image files as well as voice files. The company offers the example of an e-mail with a three-minute voice message and a 2.7MB .bmp photo attached. Without compression, the message is 6.7MB total, and would take over half an hour to send using a 28.8 Kbps modem. With compression, the message totals 167KB and can be transmitted in 47 seconds, the company claims.

Rudestam has plans for the company's future, including the expansion of the voice recognition line into the handheld market. "We will do that when it is feasible for those types of devices to employ the amount of memory and data space to handle speech recognition engines," he says. "As they move towards that, of course all our applications will migrate down, and we'll develop new ones, to take advantage of that platform. I don't see that happening necessarily in the year 2000."

Voice Pilot products are currently distributed in Canada by both Ingram Micro and Tech Data. The company is also in discussions with a third, unnamed distributor to increase their availability. Voice Pilot Continuous retails for US$109.99, while Hear-Say costs US$39.99. Hear-Look and Voice Pilot Pal will retail for US$49.99 and US$59.99 once they are released.



Go to the top.