How to use voice recognition in Windows 7
This page explains step-by-step how to set up and use Speech Recognition, the built-in voice-recognition function in Windows 7. This is an excellent introduction to the use of voice recognition for both dictation and hands-free control of a computer. However, it lacks some of the advanced functionality and integration that commercial software can offer.
Areas in this guide:
Note: The 'Ease of Access Center' has replaced 'Accessibility Options' (which was used in earlier versions of Windows) in the 'Control Panel' of Windows 7. For more information, read the Introduction to the Ease of Access Center.
Step 1: Turn on Speech Recognition
Open the 'Speech Recognition' window by clicking the 'Start' button, followed by 'Control Panel', then 'Ease of Access', then 'Speech Recognition', as shown in Fig 1.
Alternatively, open the 'Ease of Access Center' window by pressing the Windows key + U. Press Tab until 'Use the computer without a mouse or keyboard' is highlighted and then press Enter. Press Tab until 'Use Speech Recognition' is highlighted and then press Enter.
Click on 'Start Speech Recognition', as shown in Fig 2, or press Alt + S.
Step 2: Set up your microphone
If this is the first time you are running Speech Recognition, you will be prompted to set up your microphone.
To check that your microphone is working, click on 'Set up microphone', as shown in Fig 2, or press Alt + M.
Click the radio button beside the type of microphone you are using, or press Tab until it is highlighted, as shown in Fig 3. Then click the 'Next' button or press Alt + N.
You will then see a screen describing how to adjust your microphone. When you have finished this step, click the 'Next' button or press Alt + N.
Next, you will be asked to read a sentence on your screen. You'll see a bar showing the volume level move as you speak, as shown in Fig 4. Click the 'Next' button or press Alt + N.
The microphone should now be set up. Click on the 'Finish' button, or press Tab until it is highlighted and then press Enter.
Step 3: Train your computer to recognise your voice
You can now start dictating. However, the recognition will not be accurate as the computer has not yet 'learned' your voice. It's recommended that you train your computer to do this. The first training unit takes 5 minutes, and you can take further training units if you want.
In the 'Speech Recognition' window (shown in Fig 2), click on 'Train your computer to better understand you', or press Alt + R.
Read the text on the welcome screen, then click the 'Next' button or press Alt + N. Then read the training text.
You can continue by clicking on 'More Training', or press Alt + M.
When you want to end the session, click on 'Finish' or press Alt + F.
After you have done this the computer should recognise your voice.
Step 4: Customise Speech Recognition further
In the left-hand pane of the 'Speech Recognition' window, click on 'Text to Speech', or press Tab until it is highlighted and then press Enter.
If it is not already selected, click on the 'Speech Recognition' tab as shown in Fig 5, or press Tab until any of the tabs is highlighted and then use the arrow keys to select the 'Speech Recognition' tab.
In the box below the 'Recognition Profiles' header, you can select your own user voice profile, which is useful if the computer is used by different people. (Note that in the example shown in Fig 5, only one profile has been set up).
You can further customise Speech Recognition by ticking the boxes below the 'User Settings' header.
Tick the box next to 'Run Speech Recognition at startup' by clicking on it, or press Alt + R to tick it.
Tick the box next to 'Review documents and mail to improve accuracy' by clicking on it, or press Alt + O to tick it.
For users who cannot start Speech Recognition with the keyboard, mouse or other input device, tick the box next to 'Enable voice activation'.
You can also fine tune the microphone from this window.
When you are finished, click the 'OK' button or press Tab until it is highlighted and the press Enter.
Step 5: See the list of speech commands
In the 'Speech Recognition' window (shown in Fig 2), click on 'Open the Speech Reference Card', or press Alt + C, for a list of speech commands.
Although there is voice-recognition software built into Windows 7, there are also widely used packages for sale that offer excellent accuracy and even complete 'hands-free' use. For example, Nuance's Dragon NaturallySpeaking (for information, see the software producer's Dragon NaturallySpeaking webpage) and, formerly, IBM's ViaVoice (no longer supported).
These, and similar software packages, can be very powerful options for people who are not able to use a physical keyboard or mouse, or who just want to speed up their text input. The systems take time to 'train' to recognise the speaker, but with practice it is possible to input text faster than the fastest typist.
The BBC is not responsible for the content of external internet sites.