Designing Effective Speech Interfaces

Speech Recognition Suites

As recently as 1999, most speech applications had dedicated functions. Some were command-and-control systems that allowed users to control an operating system, some were dedicated dictation packages, and some were screen readers that provided text-to-speech output. If you wanted to integrate speech control into your computer system productively, you had to purchase several programs which were not always guaranteed to be compatible. However, the field is rapidly changing and the leading commercial speech programs are now sold as suites that combine all of these functions. Now a user can purchase a single package that is a command-and-control and dictation program that also provides text-to-speech output. Interestingly, they are primarily marketed and sold under the category of speech recognition or voice recognition.

Four packages currently dominate the Microsoft Windows-based home and small business marketplace: IBM's ViaVoice, Dragon Systems' NaturallySpeaking, L&H's Voice Xpress, and Philips' Free Speech. They are similar in that they are multi-modal applications that provide nearly identical functionality.

Suite Web sites

ViaVoice (www-4.ibm.com/software/speech/)

NaturallySpeaking (www.naturalspeech.com)

Voice Xpress (www.lhs.com/)

Philips (www.speech.be.Philips.com)

“I am attracted by the machine's fluidity, its ability (or so it seems) to understand. That's the lure of voice-recognition software, and using it, I feel as if I'm living in science ...