How This Book is Organized

• Chapter 2 introduces Microsoft Speech Server 2004 and focuses primarily on the Speech Application SDK (SASDK). The wizards and controls that make the SASDK so easy to utilize will be reviewed.

• Chapter 3 presents a sample telephony (voice-only) application intended to supplement an existing solution for remote salespeople. This application allows salespeople to access constantly changing customer and product data with their mobile phones.

• In Chapter 4, the Speech Application SDK is used again to create a multimodal student registration application for a community college. The application not only gives students an alternative and more efficient means of registration, but demonstrates that the college is proactive in implementing new technology.

• Chapter 5 introduces SQL Servers Analysis Services as a tool for making predictions with a large quantity of data. This chapter will identify the different types of mining models available and the reasons why each may be utilized.

• Chapter 6 presents a real-world scenario in which Analysis Services could be utilized. The scenario involves a fictional retailer in need of a more efficient distribution schedule. The chapter steps through the use of Analysis Services to create a mining model based on a randomly generated database.

• Chapter 7 introduces a rule-based application designed to handle how new fields are added to a database. A threshold value is used to determine whether the field to be added is similar to any existing fields. The chapter identifies the four phases in the implementation of this type of application.

• In Chapter 8 an example of multiple software agents is presented. It involves creating multiple applications that allow remote salespeople to access files held in the home office. It also notifies them of new sales opportunities and utilizes a Web service to retrieve data from the central office.

• The final chapter will identify possible directions for the field of computing in the next few years. It will specifically look at promising areas of research in development at Microsoft Research. It will identify additional resources you may want to refer to for more information. In addition, it will examine a few branches of AI that may be of further interest to the reader.

• All the sample applications, available in both Visual Basic and C# versions, are provided in their entirety on the book's Web site at http://ww.awprofessional.com/title/0321246268

Microsoft Research (MSR)

In 1991, under the direction of Nathan Myhrvold, Microsoft founded its own research division, Microsoft Research (http://research.microsoft.com). The idea was that Microsoft's products could only benefit from research done by the brightest minds in the country. So that is what Microsoft has been assembling for the past thirteen years. Today, MSR employs over seven hundred people in more than fifty areas, such as data mining, machine learning, natural language processing, and speech recognition. Although the main research facility is located in Redmond, Washington, there are adjunct facilities located in San Francisco, Silicon Valley, Beijing, and Cambridge, England.

MSR researchers have received many prestigious honors, including the Kyoto Prize in Advanced Technology and the Turing Award of the Association for Computing Machinery. Many of the researchers have been recruited from top universities and maintain their academic ties by routinely publishing papers for peer review. MSR is able to attract such top researchers by offering them something that not every company can freedom to work on what they want, while still having access to abundant resources.

To maintain the close connection between research and product development, MSR sponsors an annual event to demonstrate the latest advances to Microsoft developers a geek fest appropriately named "TechFest." Many of the technologies that began at MSR have made their way into current Microsoft products. Analysis Services and the Speech Application SDK are both direct results of this research.

When describing the goals of MSR, Bill Gates, chairman and chief software architect at Microsoft, stated:

"We're focusing more on research than ever. We're building the technology that will enable computers to see, listen, speak and learn so people can interact with them as naturally as they interact with other people."