Microsoft Launches Microsoft Speech Server 2004 R2, Expands Language Support
Microsoft Corp. has announced the release of Microsoft Speech Server (MSS) 2004 R2. The new version of Microsoft Speech Server, according to the company, helps lower costs, deliver improved flexibility and provide enhanced performance and reliability for enterprise customers. With MSS 2004 R2, enterprise customers can offer language support for U.S. Spanish and Canadian French at no additional cost. Customers can also deploy MSS 2004 R2 in an all-in-one server configuration, which gives them better flexibility when choosing a deployment option.
"Microsoft Speech Server 2004 R2 is the next step in delivering our promise to reduce the cost and complexity of building and deploying enterprise speech solutions for our customers,' said Rich Bray, general manager of the Speech Server group at Microsoft. 'With MSS 2004 R2, we are better meeting our customer needs by providing additional language support and new deployment options that make it easier and more affordable for companies of all sizes to benefit from speech technologies."
The new language support will address the needs of enterprises that serve Canadian and U.S. multilingual callers by enabling the enterprises to deploy speech-enabled applications designed for Canadian French, U.S. Spanish or U.S. English speakers. New automated speech recognition (ASR) and text-to-speech (TTS) engines are provided in MSS 2004 R2 at no additional cost to Microsoft Speech Server customers.
The new MSS 2004 R2 Enterprise Edition all-in-one server deployment option can significantly lower an enterprise's speech platform purchase costs, especially for small to medium-sized speech deployments servicing between 24 and 48 ports, and touch-tone solutions supporting up to 96 ports. In addition, MSS 2004 R2 Enterprise Edition continues support of distributed multiserver deployments, which are recommended for implementations requiring more than 48 ports.
Part of the Windows Server System, MSS is a Web-based, speech-enabled interactive voice response (IVR) and multimodal application solutions platform. As a Speech Application Language Tags (SALT)-based platform, Speech Server follows the Web application deployment model, using an ASP.NET-based Web server as the application server. The two main core components included with Microsoft Speech Server are the Telephony Application Services and Speech Engine Services, including ASR, TTS and prompt manager. MSS 2004 R2's all-in-one server configuration deployment gives customers greater flexibility when choosing their deployment option. In addition, customers can benefit from expanded telephony boards coverage and voice over Internet protocol (VoIP) support for MSS 2004 R2 from industry partners, including Aculab, Brooktrout Inc., Eicon Networks and Vail Systems Inc.
Intervoice Fighting Identity Theft
To fight against fraud and identity theft, some call centers are moving beyond simple security questions when verifying callers' identities. Intervoice recently announced that its Omvia Authenticator now provides customers with leading-edge voice authentication technology to increase the security of their voice applications and foster more trusted relationships with their own customers. The new application aims to make callers feel more confident and secure when using the telephone to access account information or complete transactions. This security can help boost brand loyalty, improve customer retention and reduce churn. In addition, costs can be lowered by improving self-service automation rates and reducing the amount of time it takes customer service agents to help customers complete their transactions.
Omvia Authenticator product features include:
' Advanced risk modeling to make highly accurate, real-time authentication decisions, taking into account voice biometric results, content knowledge, presence information, past behavior and profile data to make the best possible authentication decision with the highest level of accuracy and automation possible;
' The ability to be leveraged as an enterprise authentication resource across multiple applications and business needs, as it resides on a separate authentication server; and
' Usage of default enterprise database methodology for storing voiceprints and authentication profile information, allowing the voiceprint records to be managed in the customer's existing database infrastructure and maintained according to their database security and firewall rules.
The Omvia Authenticator uses expertise in multi-factor authentication with a current focus on leveraging voice biometric technology for use in large-scale, consumer-facing deployments. The underlying technology can also be used by call center agents to conduct pre-enrollment authentication to make sure that the person whose voiceprint you are enrolling is identified correctly. A simple activity such as authentication can also help large call centers save millions of dollars by automating the specific tasks of authentication activities from a live agent to an even smarter system. Beyond identification and authentication, Omvia Authenticator offers robust audit trail and proactive monitoring capabilities to meet regulatory compliance requirements. All of these process best practices are designed to increase customer trust in the integrity of their account relationships.
Definition Du Jour
Unless you've been living in a box recently, it's likely you've been hearing speech companies discussing and contrasting 'SALT' or 'VXML.' You may have even told yourself you'd eventually go and find out what exactly those terms mean. We'd like to help out. Below, you'll find two links: one to the World Wide Web Consortium (W3C) Web site and one to the site of the SALT Forum. A few minutes' reading will explain these two speech standards quite nicely.
For comprehensive information about SALT, visit www.saltforum.org/whitepapers/whitepapers.asp. To view a white paper explaining VXML, visit www.w3.org/TR/voicexml20