Windows Vista Speech Recognition Step-by-Step Guide
Updated June 2007
Summary: This brief guide includes system requirements, installation instructions, and step-by-step instructions for using speech recognition in Windows Vista. (5 printed pages.)
What is Speech Recognition?
What's New in Windows Vista Speech Recognition?
Who Should use Speech Recognition?
Benefits of Speech Recognition
Key scenarios for Speech Recognition
Logging bugs and feedback
The speech recognition in Windows Vista is state of the art. Using speech recognition in the operating system, you can dictate emails and documents and use your voice to control your favorite programs and navigate Web sites. Enabling you to create documents quickly, Windows Vista Speech Recognition helps you to work with reduced risk of repetitive stress injuries. The Windows Vista speech recognition experience has an innovative design that places you in control and enables you to become proficient at using speech recognition faster than ever before.
Speech Recognition technology allows you to control your computer by speaking. When you talk, the software will intelligently decide whether to convert your voice into text, such as to allow you to dictate documents and e-mails, or whether you have given a command that requires it to take an action.
Windows Vista Speech Recognition focuses not only on providing world class accuracy, but on providing the most usable end-to-end speech recognition experience. It addresses key issues that currently frustrate or confuse users of existing products. Windows Vista Speech Recognition provides an efficient, enjoyable way to get your tasks done with speech, including:
- A reduced learning curve, with a focus on making the system "just work."
- Redesigned user interfaces provide a simple but efficient experience so you can dictate and edit text, and correct any mistakes.
- An interactive tutorial that teaches you how to use Speech Recognition while the computer learns your voice.
- Significant improvements to accuracy with speech technology that learns as you use it.
- Control of your computer with new interfaces that let you use applications in a way that places you in control.
- Improved clarification and interpretation. When you give a command that can be interpreted multiple ways, the system will prompt you to clarify what you meant to do.
This guide is targeted at the following audiences:
- IT planners and HR departments providing assistive technology solutions for the employees who have repetitive stress injury, mobility impairments, or vision limitations
- IT planners for education departments
- Early adopters
Speech Recognition is intended to address the needs of information workers, particularly those who primarily use e-mail applications, word processing applications, and web browsers.
The following users will have less successful experiences with Windows Vista Speech Recognition:
- Users in an environment where talking to the computer would have a negative effect on others, such as in an open plan office.
- Users who often work with text that is not grammatical (programmers, spreadsheet users, computer administrators).
- Users who need a tightly integrated screen-reader and speech-recognition solution.
With Speech Recognition, you can:
- Dramatically reduce the use of the keyboard and mouse by performing tasks with speech.
- Create documents efficiently by talking to your computer.
- Control your computer using your voice, including starting or switching applications and selecting menus and buttons.
This guide discusses the simple scenario of dictating a document and giving voice commands.
The Speech Tutorial which is included as part of Windows Vista contains more scenarios that you may wish to try.
To complete this task, ensure that you meet the following requirements:
- Use a high-quality microphone. Examples include:
- A USB headset microphone
- An array microphone
- Complete the speech recognition interactive tutorial.
Note This lets the computer begin learning to recognize your voice accurately. Completing the tutorial also ensures that you learn the voice commands needed to perform tasks by voice.
Dictating a document overview
One of the most common scenarios with Speech Recognition is dictating a document, which involves the following:
- Opening your word processing application
- Dictating the body of the document
- Saving the document
Steps for dictating a document
Complete the Speech Tutorial
- In the Start Menu, choose All Programs, Accessories, Accessibility. Select the Speech Recognition option
- Proceed through the initial set-up screens to configure your microphone. The Speech tutorial will then be launched
- Complete the Speech tutorial
Create, dictate, and save a document
- Say "Start Listening."
- Say "Open WordPad."
- Say "This is a test of speech recognition period."
Note Remember to pronounce punctuation.
- To correct mistakes, say "correct" and the word that the computer typed by mistake, and then pick the correct word from the list offered by Speech Recognition, repeat the correct word again. For example, if the computer misrecognized "speech" as "peach," say "correct peach," and then choose the right word from the list, or say the word "speech" again.
- To save the document, say "File", then say "Save As."
- 6.Say "Test document."
- 7.Say "Save."
You might encounter the following issues when you use Speech Recognition.
- Speech Recognition dictation will work in a limited set of applications that support a technology called the Text Services Framework. This set of applications includes:
- Microsoft Word and Outlook
- Virtually all applications that are part of Windows Vista
- Dictation will not work in many applications. These include, but are not limited to:
- Microsoft PowerPoint
- Lotus Notes
Poor recognition of voice commands
If you are having difficulty completing a task by voice, the most likely cause is that the computer is accurately recognizing your voice, but you are saying the wrong command.
Refer to the Speech Reference Card to learn which commands are available. To access the Speech Reference Card, right click the microphone control bar.
To log bugs, use the instructions on the Microsoft Beta Web site. We are also interested in feature requests and general feedback about Speech Recognition.
You may request additional support for Speech Recognition by sending emails directly to the Speech Recognition team at firstname.lastname@example.org.
If you are a beta tester and part of the special Technology Adoption Program (TAP) beta program, you can also contact your appointed Microsoft development team member for assistance.