SpeechWorks and Kirusa Partner in Creating Advanced Multimodal Mobile Solution

BOSTON, MA and BERKELEY HEIGHTS, NJ -- SpeechWorks International Inc. (Nasdaq: SPWX), a global provider of speech recognition and text-to-speech (TTS) technologies and services, and Kirusa, a provider of platforms enabling multimodal wireless applications delivery, have partnered to develop a multimodal solution empowered by Kirusa's multimodal platform, fully integrated with SpeechWorks' speech technologies, enabling users to interact with mobile applications using a combination of speech and visual modes. The combined solution provides customers with a multimodal platform supporting applications that integrate visual and speech interfaces. On more sophisticated next-generation devices, users may use simultaneous speech, visual and gesture input with speech and visual output. On less capable devices, the solution supports multimodal application experiences through alternating speech and data in a single, contextually integrated application session. For example, in a bank transfer application, with "From Account", "To Account", and "Amount" boxes on the screen, the user may simply speak "Transfer $5000 from Account 123 to Account 456", and all three form entry boxes would be populated correctly as a result of the single utterance. Results can be delivered in voice mode, or visual, or both - giving positive confirmation of the transaction. Kirusa's MultiModal Platform (KMMP), and SpeechWorks' Speechify, OpenSpeech Recognizer, OpenSpeech Browser Platform Integration Kit are integrated to provide wireless operators and enterprises support for a variety of handheld devices, from SMS and WAP phones, to smart phones, PDAs, and other emerging device configurations. The multimodal solution also utilizes distributed speech recognition, which splits the speech processing on the device and network. The solution gives mobile operators the ability to offer multimodal applications customized to individual subscriber capabilities today. "Multimodal applications will improve user interfaces beyond speech-only and unlock the revenue potential of 2.5G and 3G wireless services and applications," said Alan Schwartz, vice president of business development and head of the Automotive and Mobile Devices Customer Focus Unit at SpeechWorks. "Leveraging our distributed speech recognition technologies in Kirusa's multimodal environment enables a multimodal experience with packet voice data rates between 3 and 5 kbps - well within limitations imposed by 2.5G networks. Mobile operators can quickly deploy one of the most advanced multimodal solutions today, without consuming excessive bandwidth or requiring a 3G network." "We are pleased to partner with SpeechWorks in delivering multimodal solutions. SpeechWorks' leadership in speech solutions, and their commitment to support multimodal platforms, fits well with Kirusa's leadership in multimodal technologies," said Inderpal Singh Mumick, Kirusa's CEO. "The joint solution realizes the shared vision between Kirusa and SpeechWorks of a broad multimodal experience that provides compelling value to the mobile user, and enhanced revenues for the mobile carriers deploying 2.5G and 3G networks."