SAN FRANCISCO, March 24, 2004 — Since the dawn of the information age, researchers have labored to build applications that will enable people to interact with computers simply by talking. But building computer systems that respond instantly and accurately to the spoken word has proven to be much more difficult than most peopleanticipated. Because of the sheer complexity and variety of human speech, only the largest organizations have been able to afford the time and expense required to build sophisticated speech-enabled applications. As a result, speech recognition software that can offer much more than “Press one or say one” functionality has been something of a rarity.
With the launch of Microsoft Speech Server 2004, the world of speech recognition software is poised to undergo an important change. Introduced by Microsoft Chairman and Chief Software Architect Bill Gates during his opening keynote address at the joint VSLive! and AVIOS~SpeechTEK event in San Francisco today, Microsoft Speech Server is a standards-based, integrated speech platform that will deliver the business benefits of speech-enabled solutions to both large-scale enterprises and midsize companies. The release of Microsoft Speech Server marks the shift of speech recognition from niche technology to mainstream application that can help companies reduce costs, raise productivity, increase customer satisfaction and take advantage of new opportunities for generating revenue.
“Microsoft Speech Server provides an extremely cost-effective and flexible platform that integrates telephony and the Web, while offering great security and reliability, along with speech recognition that is second to none,” says Xuedong Huang, general manager for the Microsoft Speech Server product group. “Speech has always been an important part of Microsoft’s vision for seamless, pervasive computing. If you think about the fact that there are 1.2 billion telephones in the world, Speech Server is really going to open the door to a whole new range of solutions that it will make it possible to access the information you want even when you don’t necessarily have access to a computer.”
The Beginning of Speech as a Mainstream Technology
First made available last July with the release of the beta 1 version, Microsoft Speech Server is already playing a significant role in helping independent software development companies create mainstream speech-enabled solutions. Currently, more than 60 companies are working with Microsoft to develop packaged speech applications and services through the Microsoft Speech Partner Program. Companies such as Jet Blue and Grange Insurance, as well as the New York City Department of Education, are taking advantage of speech-enabled solutions built on Microsoft Speech Server to streamline operations and deliver better customer service.
“It’s one thing to be able to say ‘yes’ or ‘no’ or “one, two, three,'” says Rick Souler, director of marketing at Viecore, Inc., a leading systems integration firm that specializes in creating self-service solutions for customer contact centers. “But natural language is a whole new thing. With Microsoft Speech Server, I think we’re really right at the beginning of speech as a mainstream application, and you’re going to see some exciting new solutions become available very quickly.”
According to Souler, Viecore is utilizing Microsoft Speech Server to develop speech-enabled solutions for enterprise and midsize companies in industries ranging from financial services to utilities, healthcare, insurance and retail. Currently, Viecore is putting the finishing touches on an auto-attendant that will enable people who call a central switchboard to simply say the name of an employee, department or service in order to be automatically connected.
Solutions like Viecore’s auto-attendant are just the beginning. Speech-enabled applications based on Microsoft Speech Services under development by Microsoft partners include solutions for customer relationship management, call centers, financial service automation and prescription fulfillment. At Accenture, the global management consulting and technology services company, CRM professionals and the technology research and development organization have worked together to create two prototypes that take advantage of the Microsoft Speech Server platform.
Both prototypes focus on using speech to reduce the cost of customer interaction and improve the customer experience. The first integrates Microsoft Speech Server with Accenture’s Customer Insight engine, resulting in an application that enables customers to move quickly through a streamlined, automated experience using voice prompts to explore and order new services. The second uses speech recognition capabilities to enable on-site sales professionals to answer questions and help customers make purchase decisions.
An Integrated Speech Platform Built on the Windows Server System
A new member of the Microsoft Windows Server System family of products, Microsoft Speech Server was created to deliver a flexible, integrated speech platform that delivers the business value of speech in a truly cost effective manner. Built on the Microsoft Windows Server 2003 operating system, Microsoft Speech Server provides high reliability and availability, increased performance and scalability, and a wide range of advanced security features.
Microsoft Speech Server also includes a set of Speech Application Language Tags (SALT) development tools that take advantage of the Microsoft Visual Studio .NET Web-programming model. Together Microsoft Speech Server and the speech development tools form a platform for building speech applications that offers low total cost of ownership, the flexibility to meet real-world business needs and the ability to interoperate with existing applications and solutions.
In addition, because Microsoft Speech Server combines Web technology with speech-processing services and telephony capabilities in a single system, developers can create applications that provide either voice-only or multimodal features. With Microsoft Speech Server, companies can deploy speech-enabled solutions that can be accessed by telephone, cell phone, Pocket PC, Tablet PC and other devices.
Microsoft Speech Server is available in Standard and Enterprise editions. Microsoft Speech Server Standard Edition was designed for single-server configurations and is focused on small and medium-sized deployments. For large enterprises, Microsoft Speech Server Enterprise Edition supports speech deployments that require highly scalable and distributed configurations. With the Enterprise edition, customers can scale out by adding as many nodes as needed to meet their port capacity requirements.
Microsoft Partner Companies Deliver Key Enabling Technologies
Along with the Bill Gates’ speech announcing the general release of Microsoft Speech Server, SpeechTEK Spring 2004 includes demonstrations of speech technologies and solutions built by companies that have worked with Microsoft during the development of Microsoft Speech Server and as part of the Microsoft Speech Partner Program. Among the companies at SpeechTEK Spring 2004 are such leading communication technology companies as Intel, Intervoice, ScanSoft, Envox Worldwide, Voice Automation, Gold Systems, Redmond Software, CYGCOM and Tata Consultancy Services.
Intel, the world’s largest chip maker, is also a leading manufacturer of networking and communications products, and the Intel Communications Group has worked closely with Microsoft to develop a reference architecture that enables Microsoft Speech Server to interoperate with existing telephone networks. Intel is providing two key pieces of that reference architecture: the telephony board, which is a piece of hardware built by Intel that allows voice-enabled solutions based on Microsoft Speech Server to connect directly with existing telephone systems, and the Telephony Interface Manager, or TIM, which is a layer of software that simplifies integration of the telephony board with Speech Server.
“Speech is a really compelling technology because it provides such a natural way to interact with computer systems,” says Intel Communications Group Marketing Manager Peter Gavalakis. “Until now, the cost, complexity, the lack of standards and the lack of interoperability have been significant impediments to the widespread adoption of speech technology. Microsoft Speech Server has the potential to change that, and I think Microsoft has done an outstanding job of making it easy to build great applications by providing excellent tools, delivering a really reliable product and ensuring that there is a strong ecosystem of partners who understand the technology.”
Intervoice, a Dallas-based company that has been developing voice solutions since 1983, will also be an important presence at SpeechTEK. Intervoice has also worked with Microsoft to develop a Telephony Interface Manager. At SpeechTEK, the company will describe how the Intervoice TIM can help turn Microsoft Speech Server into a robust, feature-rich voice automation platform that integrates into virtually any call center or telephony environment.
In addition to its work on the TIM, Intervoice has also deployed prototype applications based on Microsoft Speech Server for two major customers: the clothing retailer Talbots, and the banking and investment firm Huntington Bancshares.
According to George Platt, senior vice president of marketing and corporate strategy, one important impact of the release of Microsoft Speech Server is that the range of voice-enable solutions will grow, which will expand the market for speech applications significantly.
“Call centers have been the traditional market for speech technology,” says Platt. “The launch of Microsoft Speech Server gives speech technology a lot of visibility and credibility, and early on you’ll see new solutions for call centers of all kinds, from financial services to insurance companies and retailers. But other kinds of applications will soon follow, such as human resources solutions that provide employee self-service. That’s really going to help this industry grow.”
The Growing Ecosystem of Speech Technology Companies
ScanSoft, a global leader in speech technologies, applications and services, is another Microsoft partner that is excited about the potential for growth in the market.
“Microsoft Speech Server is going make it possible for many midsize companies to tackle speech projects for the first time,” says Dave Nuckols, Scansoft’s director of the Microsoft strategic relationship. “Until now the development process for custom applications has been so costly that only companies with high call volumes could justify the expense. Speech Server makes speech solutions more cost-effective to build, deploy and run, allowing smaller organizations to capture the benefits of speech automation more broadly and more deeply.”
According to Nuckols, ScanSoft has worked with Microsoft on a number of important technologies for Microsoft Speech Server, including the text-to-speech and automatic speech recognition software that is part of the Speech Server Software Development Kit. ScanSoft is also a good example of the strong ecosystem of partners that Microsoft has created to help technology developers build and deploy speech solutions. ScanSoft has developed and deployed scalable, high-quality speech applications around the world using proven processes and best practices in user interface design to ensure the highest levels of caller satisfaction and ROI for their customers.
“As part of this ecosystem around Speech Server, we’re working with other Microsoft partners as they build their speech practices,” says Nuckols. “The combination of our expertise in areas like application development, core speech technologies and user interface design, and the .NET expertise that Microsoft partners can bring to the table makes for a really strong package for companies looking to take advantage of speech automation.”
At SpeechTEK, ScanSoft will demonstrate its Change of Address packaged application that automates the way call centers collect address information. Change of Address can be used to start or stop service for a utility or to forward a magazine subscription to a new address. Today, change-of-address information has to be collected by call-center operators and manually entered into a call-center database. The ScanSoft solution will allow callers to simply say their address information after hearing a prompt, saving the call center significant time and expense.
Envox Worldwide, a global provider of voice-enhanced communications solutions based in Stockholm and Boston, has been working with Microsoft to develop open, standards-based speech solutions since July, 2003. At SpeechTEK Spring 2004, Envox Worldwide is discussing its plans for supporting the SALT standard. Those plans include developing a line of prepackaged, self-service applications for retail stores and pharmacies that run on Microsoft Speech Server. In addition, Envox Worldwide has launched a SALT practice as part of its professional services organization that will provide development, call flow design, systems integration and speech user interface design for SALT-based solutions.
Envox Worldwide will also unveil its first SALT application. Created for pharmacies, the new solution uses Microsoft Speech Server to automate the process for refilling prescriptions by providing a well-tested call flow and speech user interface. According to Envox, the solution will enable pharmacies to use voice automation to deliver higher levels of service to their customers at a price point that was not possible before Microsoft entered the market.
Speech Partner Program Companies Unveil Speech-Enabled Solutions
Another important participant in the Microsoft Speech Partner Program that will be at SpeechTEK is Voice Automation, Inc., of Westford, Massachusetts. Last October, Microsoft and Voice Automation announced a strategic relationship aimed at voice-enabling Microsoft Business Solutions CRM. Voice Automation for Microsoft CRM allows remote workers to access and update enterprise sales data without a keyboard. With the Voice Automation solution, employees who are on the road will be able to use a phone to provide spoken commands that allow them to navigate through Microsoft CRM.
Voice Automation for Microsoft CRM offers simplified, secure login from either a land or mobile phone. At SpeechTEK, Voice Automation will showcase key features of the solution including the ability to retrieve and respond to e-mail; access contact information details such as telephone numbers and driving directions; and update lead status.
Boulder, Colorado-based Gold Systems, Inc. will demonstrate two solutions based on Microsoft Speech Server. The first is aimed at the insurance market, and includes a set of pre-packaged speech modules that provide self-service access for policyholders. The reusable modules, which can be easily customized to meet the specific needs of individual companies, can significantly reduce call center costs and increase customer satisfaction. During the product demonstration, Laura Farrelly, Gold Systems vice president of product management, will preview account status, claim status and physician locator insurance modules.
Another key participant in the Microsoft Speech Server Program that will demonstrate a new product at SpeechTEK is CYGCOM. Based in Markham, Ontario, Canada, CYGCOM will showcase its SpeechWatch Server, which monitors critical elements of system performance. Built for enterprise deployment, SpeechWatch Server provides IT professionals with reliability in a fully supported platform for distributed speech applications.
Redmond Software will also be at SpeechTEK, where it will unveil its speech-driven auto attendant. The system, which can accurately and efficiently transfer inbound calls to the appropriate person or department within a company without operator intervention, includes tools to get names automatically from Microsoft Active Directory, Microsoft SQL Server or from text files.
Gustavo Berdinas, CEO of Redmond Software, is enthusiastic about the support the Microsoft Speech Server partner network has provided. “We’re a smaller company,” he says. “But because we teamed up with several other Speech Server partners we are able to deliver this new class to of products to our customers at a low price point without compromising on quality and service”
Early Speech Server Adopter Gets a Jump on the Market
Tata Consultancy Services (TCS) will also be demonstrating Microsoft Speech Server-based solutions at SpeechTEK. One of the world’s leading global consulting, IT solutions and services enterprises with more than 24,000 consultants, India-based TCS was the first Microsoft technology partner to implement Microsoft Speech Server into a live customer environment.
Built for Grange Insurance, the speech-enabled solution was created to enable callers to retrieve billing details of a policy 24 hours a day, seven days a week. When it was deployed, the system was built to handle 20 percent of policy inquiry calls, with the expectation that it would scale over time to handle 80 percent of those calls.
According to TCS development center manager Nitin Hanjakar, the success of the Grange Insurance application has led TCS to explore a wide range of potential applications based on Microsoft Speech Server. At SpeechTEK, he will showcase TCS’s Round the Clock Billing Inquiry application, which allows callers to get policy or mortgage billing details over the phone. For insurance policies, that information can include payment due date, last payment date, policy status and payment status. For mortgages, information can include home coverage amount, deductible, loan number and physical address of home. The solution not only allows customers to access key information 24 hours a day, but it reduces customer service representative workload significantly, enabling them to focus on cross-selling and up-selling opportunities.
Other Microsoft Speech Partner Program members slated to unveil speech-enabled solutions built on Microsoft Speech Server include Edify Corp. and GM Voices Inc. Partners that are providing system integration services and customer application development services include Enterprise Integration Group, HeyAnita, HTI Voice Solutions, InfoSys and Wipro Technologies. For a complete list of partners, see Microsoft Speech Server Partner Information under Related Links at right.