Philips Speech Processing Launches SpeechMania 7.0 for Natural Language Voice Recognition in Fixed and Mobile Networks

Business Wire, Feb 6, 2001

Business/Technology Editors

DALLAS--(BUSINESS WIRE)--Feb. 6, 2001

Philips Speech Processing today introduced SpeechMania(R) 7.0, a much enhanced version of its highly successful speech recognition platform.

Based on natural language understanding and TrueDialogue(TM) technology, SpeechMania 7.0 allows people to use normal human conversation when interacting with fully automated telephone-based information and transaction systems. SpeechMania systems recognize user requests by identifying the relevant information and interpreting it in the context of the user's statement.

SpeechMania 7.0 includes powerful new features, including mixed acoustic model support (simultaneous phoneme-based and whole-word based speech analysis), channel independent speech recognition, and support for 22 different languages. This enhanced speech recognition platform is suitable for implementing a wide range of high-end dialogue-based information and transaction services, and is especially suited to telephony applications accessed via PSTN (Public Switched Telephone Network) or cellular networks.

"The new features in SpeechMania 7.0 make it especially suitable for network based and customer care applications where the service content requires regular updating," says Matthias Pankert, director of Product Strategy, Planning and R&D at Philips Speech Processing. "We are particularly proud of SpeechMania 7.0, which is by far the most sophisticated speech recognition platform on the market, enabling developers to rapidly and cost-effectively create new applications or enhance the performance and functionality of existing applications."

In addition to introducing SpeechMania 7.0, Philips Speech Processing has also introduced two new add-ons for the software. The Generic Database Interface simplifies data management and performance improvement, supporting a wide variety of commercial databases (such as ORACLE and ODBC-DB2) and offering dedicated functions for data queries and data manipulations.

The new LearnWord Transaction add-on enables end-users to personalize the speech interface by adding, modifying or deleting words and phrases contained in a user-specific vocabulary library. This custom library is then dynamically loaded into the recognizer and Speech Understanding Module whenever the user accesses the system.

SpeechMania allows developers to create applications quickly and easily using a high-level dialogue description language (HDDL), and a comprehensive developers' toolkit that supports each stage of the development process. Its new mixed acoustic model support makes it more flexible than ever before in supporting a wide range of applications. Other features aimed at easing the development of new applications include transaction improvements that simplify the integration of SpeechMania into other programs and also improve debugging/reliability testing, plus full SNMP support for remote maintenance and diagnostics in Window NT based installations.

The SpeechMania platform has already been deployed by numerous companies around the world for a wide range of different applications. These include:

-- the voice portal for Italian mobile phone operator Omnitel, which currently
provides 70 different information services to its customers using the
SpeechMania platform

-- the world's first Mandarin language voice portal, developed for KG Telecom,
one of the leaders in Taiwan's emerging telecommunications market

About Philips Speech Processing

Philips Speech Processing, a business unit of Royal Philips Electronics, is a pioneer and global leader in speech recognition, natural dialogue and language understanding technologies, with more than 40 years experience in the development and marketing of speech products. Philips offers a broad portfolio of speech technology solutions to the telecommunications industry. As a developer and provider of speech technologies in multiple languages, Philips has the largest installed base of speech recognition and natural language understanding systems in Europe, and is a large and growing provider of speech technology in the Americas and other regions of the world. Installations include call center services automation, voice portals, telephony applications and Internet applications. Visit www.speech.philips.com/telephony for additional information.

About Philips

Royal Philips Electronics of the Netherlands is one of the world's biggest electronics companies and Europe's largest, with sales of EUR 31.5 billion in 1999. It is a global leader in colour television sets, lighting, electric shavers, colour picture tubes for televisions and monitors, and one-chip TV products. Its 239,370 employees in more than 60 countries are active in the areas of lighting, consumer electronics, domestic appliances, components, semiconductors, medical systems. Philips is quoted on the New York (NYSE:PHG), London, Frankfurt, Amsterdam and other stock exchanges. News from Philips is located at www.news.philips.com

Editor's Note: Philips has created the following rapid-response, toll-free number for all editorial inquiries: 800/210-9605. Digital images are available upon request. Broadcasters: B-roll on Philips is available from Bulletin International at 212/899-5450 or 415/252-7106.