SpeechWorks Acquires SpeakEZ Verification Technology and Related Patents

BOSTON, MA - SpeechWorks International Inc. (Nasdaq: SPWX), a developer of speech recognition and text-to-speech (TTS) technologies and services, announced that it has acquired the SpeakEZ Voice Print patented technology from T-NETIX Inc. (Nasdaq: TNTX). SpeechWorks has previously licensed SpeakEZ for its SpeechSecure OpenSpeech DialogModule. This acquisition will allow SpeechWorks to further develop the SpeechSecure product for both network-based services and embedded applications and to support customer requests for additional features such as random digit collection and text independent recognition. A key advantage of the SpeakEZ technology is its process of combining several different algorithms to generate and verify voiceprints. This 'fusion logic' process has proven a higher rate of accuracy than that achieved by single algorithms. In addition, SpeakEZ technology, allows customers more flexibility in choosing a speaker verification solution because it is not dependent on automated speech recognition software. SpeechWorks will own two patents associated with the product and license an additional three patents from Rutgers University. Dr. Kevin Farrell, the primary developer for SpeakEZ, will join SpeechWorks to lead the future development of the product. T-NETIX will continue to serve the corrections industry through a license with SpeechWorks. "This acquisition completes a strategic process we undertook over two years ago to offer our customers speech recognition, TTS and speaker verification for both networked and embedded environments," said Stuart R. Patterson, CEO of SpeechWorks. "From experience, we know that this industry-leading technology, backed by our solutions services team, will help us respond to the increased global demand for speaker verification and identification solutions." Over the past twelve months, SpeechWorks has witnessed a growing interest in its speaker verification solution due to greater security concerns and federal legislation such as the Health Insurance Portability and Accountability Act (HIPAA). HIPAA requires healthcare organizations to implement new security and privacy standards such as 'entity authentication' in order to protect personal health information. Speaker verification technology has also been beneficial to call centers that can now automate caller authentication, personal identification number (PIN) reset, and other labor-intensive processes, resulting in a compelling return on investment. "This acquisition is a clear example of the growing role that voice biometrics are playing in speech-processing solutions for enterprises and carriers," said Judith Markowitz, president and founder of J. Markowitz, Consultants. "By bringing SpeakEZ in-house, SpeechWorks can better exploit the full potential of the technology and patents, to address new opportunities in the speaker verification market." SpeechSecure OpenSpeech DialogModule software uses the SpeakEZ voice biometric technology to verify a caller's identity based on the characteristics of his or her unique vocal patterns. SpeechSecure software provides an added layer of security in applications by associating a voiceprint with information that the caller is prompted to provide, such as an account number, password or other challenge question. In addition, SpeechSecure OpenSpeech DialogModule software represents a speaker verification module that can be used within VoiceXML applications without proprietary extensions.