Nuance Announces Latest TTS Software

MENLO PARK, CA - Nuance (Nasdaq:NUAN) announced Nuance Vocalizer 3.0, text-to-speech (TTS) software. Designed to improve caller satisfaction while reducing application development time and costs, Nuance Vocalizer 3.0 is intended for use in automated speech solutions including directory assistance, information access and retrieval, account management, voice activated dialing and email access. Nuance Vocalizer 3.0 features "out-of-the-box" coverage of application prompts, support for U.S. names and addresses, technologies to emulate a live agent, and other advanced capabilities. Customers and partners who will be able to leverage the advantages of Vocalizer 3.0 include companies such as Avaya, EarthLink powered by Audiopoint, Intervoice, and ISx's Sprint LTD project. "Nuance Vocalizer 3.0 is the first new Nuance software engine released in 2003," said Ron Croen, president, CEO and chairman, Nuance. "Partners and customers alike can take advantage of the technological benefits and advanced features. This is just one of the core software improvements we are planning for this year." Increasing Caller Satisfaction While Decreasing Development Costs To date, TTS software has primarily been used as part of speech applications that deliver frequently changing information that is difficult and expensive to handle with recorded prompts. By converting written text into spoken words, TTS software has allowed companies to automate the delivery of frequently changing information over the phone. For example, a large telecom service provider uses TTS software as part of an integrated speech recognition system to automate access to directory assistance information. Speech recognition is used to understand a caller's request, recorded prompts are used to communicate information that rarely changes, and TTS software is used to speak dynamic information such as the name and address for the listing - information which is stored as text in a database. Nuance Vocalizer 3.0 has been designed to not only help ensure high levels of caller satisfaction by communicating changing information, but also to decrease the application development costs and time associated with producing recorded prompts. Professional prompt recording can cost over $800 per hour, with an application taking weeks to record and edit in a professional studio. Therefore, a reduction in the number of recorded prompts is very desirable. Even a relatively small reduction in the number of recorded prompts can have a significant impact on the application cost and time to deploy. Key features of Nuance Vocalizer 3.0 include: -- Out of the box coverage of key prompts and dynamic information: Based on analysis of deployed speech applications, Nuance has improved coverage of the most frequently spoken words and phrases used in certain key applications. For example, Nuance Vocalizer 3.0 can completely cover many directory assistance applications, replacing the prompts that are typically recorded and also clearly rendering the dynamic information. Pre-loaded stock names and other optimizations enable the software to read news and financial information while pre-loaded airline, airport, flight numbers and city names simplify access to travel information. -- Extensive coverage of U.S. names and addresses: Nuance Vocalizer 3.0's extensive dictionary includes more than 97% of people names based on U.S. census data. It also handles business names, street names, and addresses naturally. These capabilities are essential for applications that require callers to provide or update personal information such as billing addresses. It is also necessary for applications such as voice-activated dialing, driving directions and branch and store locators. -- Emulating a Live Agent: Nuance Vocalizer 3.0 emulates the conversational style of an agent, even pausing at the appropriate times for purposes of emphasis or clarity. To the listener, it almost sounds as if Vocalizer periodically "takes a breath." Nuance Vocalizer 3.0 pronunciation improvements also help the caller feel as if they are speaking with a real person. In addition, the intonation, speed and volume of Nuance Vocalizer 3.0 can be adjusted, if needed, since Vocalizer 3.0 utilizes the latest standards in the industry -- the W3C's VoiceXML-supported speech synthesis markup language (SSML). -- Nuance Speak As One: Nuance's Speak As One offering is a set of technologies and services that enable the same voice to be used for TTS and those prompts that must be recorded, as well as tuning the transitions between the two. This results in a single, high-quality voice and "personality" throughout the system and a more conversational and satisfying caller experience. -- Rapid Creation of Custom Voices: Many companies use a specific voice talent or celebrity as their spokesperson and want to use that same voice in their speech applications. Within months, such special voices can be added to Nuance Vocalizer 3.0 to create a caller experience that leverages this unique personality. For example, a company employing a Hollywood star in an ad campaign can bring that actor's voice not only to the recorded voice prompts of a customer service application, but also to the TTS software of that same application.