Based on Bell Labs' research, the new Lucent Speech Application Platform allows users to develop Windows-based applications for speech-enabled products, the company said. The new software is available to developers for $495.
The new Lucent software, the result of more than 20 years of Bell Labs' speech research, according to the company, integrates automatic speech recognition and text-to-speech synthesis technologies for computer applications that work with a microphone or over the telephone.
Distributed on CD-ROM with multilingual engines ASR and TTS, the Lucent Speech Application Platform allows software developers to create such applications as email readers, voice-controlled Web browsers, and automatic news readers, the company said.
The speech technology, previously available only for large-scale phone network deployment as AT&T operator assistance, voice dialing, and financial services call automation, will now be available as a PC software product.
Intel executives said the text-to-speech technology is a key application for handheld devices.
"The Lucent TTS engine is ideal for this new business application area because it enables the highest voice quality and message density application area for handheld devices," said Skip Matthews, product development manager for Intel. "Speech can now move beyond the desktop."
Lucent and Intel are working to expand the market for applications of the TTS engine through a cooperative project to develop and refine for production of "away-from-the-office" business applications that will complement the capabilities of both companies, according to Lucent executives.
Lucent hopes to take advantage of its 30,000 deployed speech recognition channels and 800 corporate customers using ASR and TTS technologies for such applications as reverse directory services, credit card billing, "natural-language" call routing, and voice dialing to take advantage of the new PC speech platform with Bell Labs research.
The Lucent ASR engine supports speaker-independent, continuous speech recognition, and run-time vocabulary change. The engine can reject out-of-vocabulary utterances and offers an optimized grammar compiler. The engine uses 32MB of RAM and requires a 166-MHz Pentium Pro. It is currently available in English and Spanish, with other languages planned, the company said.
The TTS engine, on the other hand, supports American English as well as German, Spanish (Mexican), and will soon handle French, Spanish (Castilian), Italian, Russian, Japanese, and Chinese (Mandarin).
Voice recognition software allows users to dictate, rather than type, documents, emails, and basic commands. It also provides a natural way for users to interact with their computers.
Though it has been touted as the epoch-making application for the PC industry by analysts and users alike, it has yet to catch on in large numbers. Chip giant Intel and computer makers have high hopes, as there have been few compelling mainstream productivity applications that might justify the purchase of a PC with the most powerful Pentium II processors.
(Intel is an investor in CNET: The Computer Network, publisher of News.com.)