InformationWeek

Related links from our sister publications:

Windows Speech Recognition/PIM Combo-You Talk, It Does the Rest Windows Speech Recognition Is the Talk of the Town Computer Reseller News Speech Technology Takes Limelight

fair number of voice-recognition utilities execute Windows commands based on voice input. A few dictation-oriented speech programs are designed to turn users' speech into text in desktop programs such as word processors. Because of the demanding nature of blending the two tasks, no one has come out with a product that does both in a convincing way.

Lernout & Hauspie Speech Products NV is aiming to deliver that combination with version 4 of L&H Voice Xpress Professional. To a surprising degree, it succeeds. As with all recognition technologies, there's room for improvement--and users expecting a magic bullet that works perfectly will still be disappointed. Yet, this product shows some noteworthy advances, especially in the speed with which a new user can get started delivering an actual end product.

Most speech-recognition programs require significant training to be truly effective. Dragon Systems Inc.'s NaturallySpeaking starts with a 15- to 30-minute drill with entertaining reading. While this is not onerous, many users might reject the effort as too much work. The L&H product tries to overcome that behavioral barrier by squeezing the necessary training time to less than five minutes. There is, however, a necessary, ongoing cycle of refining the recognition, just as there has always been. And in its target balance between dictation and commands, the product seems to do a better job with dictation.

Version 4 adds a number of features. It sports integration with Microsoft Office 2000 and Internet Explorer 5.0. The integration works quite well. The toolbar Voice Xpress puts up sits above the application's title bar and provides the level of control you need. As you dictate text or speak commands, Voice Xpress sits in the background and executes your dictation or instructions. By using the buttons in the toolbar, you can specify whether you're issuing commands or dictating text. That helps focus the recognition to be most effective at interpreting what you're saying.

There is a set of Talking Tools-- five small desktop accessories that take advantage of the voice interface. These are of uneven value. The voice calculator is effective, but the clock and the address book are more trouble than they're worth. To me, they look like they could be good training aids for users who have a difficult time using the product in its major applications.

Voice Xpress has an expanded vocabulary and a newly optimized language model. For IT's convenience, there's an auto-recognition feature that examines the resident hardware and installs the highest level of language model that hardware can support.

The product also has a new help tool that includes user tips for improving speech-recognition accuracy. But don't ignore the incompatibility between people's beliefs about what the technology should do and what it actually does. If you go through the short training period Voice Xpress demands, follow all the tips, and let it examine all your existing files to optimize its understanding of the vocabulary you use, it will work very well.

With about eight hours of use, Voice Xpress worked well enough for me to prefer it to typing. With 12 hours of use (work and training) it's a hands-down winner.

Voice Xpress still requires more help from me than I'd like recognizing Windows and application commands. Even going to the Voice Xpress toolbar and clicking the button that tells the utility to expect a command doesn't guarantee it will recognize my command every time.

In terms of desktop applications, Voice Xpress works best with Microsoft Word and PowerPoint, both text-intensive processes. I struggled a little bit to have it work with my spreadsheet, and while it occasionally pulled the correct set of format and numbers ($1,287, for example) out of a string of spoken input, this complex task requires more training. Users who work extensively with spreadsheets may find the payback time quick enough.

Documentation is more than adequate. Still, the set of tools used to take advantage of everything that Voice Xpress makes available is pretty big and it requires a lot of user choices. Documentation makes a good effort to bring it all together as one coherent process, but there's room for improvement here, too.

There are three groups of users who will get the most value from this intelligent product. The first is executives or others who were too important to learn to type. Another group is people who suffer from repetitive-motion injuries or are at risk for them. The third, possibly the largest and certainly the most difficult to judge, is folks who will be patient enough to reap the rewards of proper setup and system training.