Continuous Dictation: New Voice Xpress Released

Lernout & Hauspie has released an entirely new edition of Voice Xpress. We review here "Professional" version 2.02, which is similar to lower editions of the same vintage, the "Standard" and "Advanced" editions, available in price ranges of about $50, $80, and $130. The standard version is minimal. The advanced version allows one to work within Microsoft Word, and contains additional vocabulary. The advanced version also has a still larger vocabulary and allows the use of macros and transcription from recorded material. This version is almost like a competely new product compared with the original release. The new version has many positive helpful new features. The recognition accuracy is greatly improved, and can recognize language spoken at an incredibly fast rate. This is useful for rapid dictation, as well as transcription from recorded material. At times, the recognition rate is remarkably high. In addition, this new system does indeed appear to take more "natural speech." Moreover, there is an optional feature which allows the recognition rate to be set as faster and less accurate, or, slower and more accurate. Slower and more accurate is often more desirable, from my viewpoint, since one can dictate at length, and then go back and make corrections after the recognition process has taken place. Faster recognition presumably comes with a fast computer processor. It will run with 48MB of RAM in Windows 95 but more memory is always desirable. This version is much more stable and crash-free compared with the original release. The program includes a pleasant speech synthesizer for reading back (in a male computer voice) recognized transcribed material, as well as other text from any source. This feature could be used to read email. A "command browser" allows one to see what commands are available at any given point, and is a useful teaching and reference feature that is easily available on the task bar. Another very nice feature of this system is the ability to turn on a small control/task bar at the top of the screen, which allows one to dictate into virtually every Windows application we have tested, and even dictate into a DOS window! Moreover, the correction window feature available in the Windows Applications can be turned on whenever text is selected, a potentially very useful feature for correction. The correct word can be spelled, using the standard alphabet, with the letters dictated into the correction box. Alternative word choices are listed in the correction dialogue box, and may be chosen by saying "take one, take two, etc." The program comes with a dictation word processing window available, similar to WordPad, with very nice features such as an on/off button for the microphone, which is conveniently placed near the top of the text area so that it can easily be turned on and off while writing. In addition, the dictation system integrates well into Microsoft Word, modifying its control bar with special dictation features. This is a very rich, full-featured system with tremendous depth. It can be optimized and customized in many ways. Particularly notable is the ability to enter into a dictation only mode, or a command mode, or a spell mode, or the "normal" mode, which allows the simultaneous use of commands and dictation. The program comes with an audiovisual teaching module and plenty of online help. I expect that a computer user could master it with so much help available in the program. As far as I can tell, the program at this point lacks the ability to play back the speaker's voice, however no doubt this feature will be incorporated into some future versions. In at least the professional version, the program does contain the ability to transcribe recorded material from tape and digital recorders. This product is simple and elegant in some of its user interface features, and yet remarkably complex in its options and depth. There are of course some rough edges which, no doubt, will be smoothed out with time. This is certainly a program which a user or organization might consider for its dictation speech recognition work. It can be used in coordination with some other speech recognition programs, however there are often conflicts between these programs and potentially one may disable some features of another. All in all, I can recommend this program as an outstanding step forward in the world of dictation speech recognition. Like all other programs, one needs to see how it performs with continued learning over time. This is the ultimate test of any dictation speech recognition system: how well does it learn over time, how well can the recognition accuracy be boosted with continued tweaking, and how fully can it be customized to the individual user's needs.

Peter Fleming, a speech recognition consultant, may be reached at aris@world.std.com, or (617) 923-9356.

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues