Published:2017
By Jeremy Moskowitz, Windows XP Expert Zone Community Columnist
Related Links
|
Editor’s Note: Past articles by members of the online community are archived for your use. The information may become outdated as technology changes. For the most current information, please search the Web site or post a question in the newsgroups.
Personal computing in the past decade has exceeded our expectations in many ways.But in a few specific ways, PC technology struggles to fulfill expectations set way back in the 1950s and 1960s.
One example is speech recognition. Forty years ago, scientists predicted the new millennium would see us all going home at night, tucking the kids in, putting the cat out, and speaking aloud to a PC: “Computer, turn out the lights and close the garage doors. Wake me at 7 with my favorite song. Make my coffee. And call my boss by 8 to remind him of our meeting.
“Goodnight computer.”
“Goodnight sir.”
Sorry. Windows XP can’t make your computer do that. However, once your expectations are adjusted for present day reality, you might be impressed with how much you really can do by talking to your computer.
The speech recognition capabilities in Windows XP, especially when combined with Office XP productivity software, can enhance computing in such areas as gaming, data entry, or editing. Further, third-party software vendors are taking the speech recognition capabilities of your PC into new areas, including things like home automation and telephony.
Microsoft has been working on speech recognition technology as a product for several years. You can learn more about the current state of speech recognition products on the Microsoft .Net Speech Technologies Web site. At the same time, Microsoft continues to invest in speech recognition research. Groups in Redmond and in China are working to solve some of the shortcomings that still nag speech recognition. One research project, known as the Dr. Who project, aims to solve the problem of noise and allow your computer to recognize your voice if you’re not in a quiet office, but outside, in a car, or in a crowded restaurant.
So what can you do now?
Tapping into the speech recognition technology within Windows XP can be a little tricky, but after reading this, you’ll be a pro.
The magic of speech recognition happens when three pieces come together. First, you need Windows XP Service Pack 1. Next, you’ll need the Microsoft Speech Recognition Engine v5.0. Finally, you need an application that’s ready to accept your speech input, these include Notepad, Outlook Express, and others in Windows XP.
There is no Speech Recognition Engine (SRE) built directly in to Windows XP. You need to install a compatible engine, and in most cases, you’ll install it from one of two sources.
|
•
|
The Microsoft SRE for Windows XP is available in Office XP applications, such as Microsoft Word 2002, or Microsoft Excel 2002. If you have Office XP, or one of these programs on your computer, then you probably already have access to the SRE. You simply need to install the SRE as explained below.
|
|
•
|
A third option is available for advanced users. The SRE is provided for free as part of the Microsoft Speech Software Development Kit 5.1. Microsoft provides no technical support for this software and it is not generally recommended for end users.
|
Before you go any further, we need to take a minute or two to talk about the most important component in your Speech Recognition system: the microphone. You will need to get the right microphone for the job. This isn’t very easy.
If you head on over to your local computer store, you’ll see lots of headphones with microphones attached to them. Even the packaging says things like “Great for speech recognition.” But this is generally not true, for multiple reasons.
First off, the quality of these microphones is questionable, which leads to decreased word recognition. More importantly, none of these microphones is “powered.” Why is this important? Because either the microphone, or the sound card’s microphone input must be powered or speech recognition won’t work. Few sound cards have powered microphone inputs, and almost none of the microphones is powered. So, how do you know if your microphone is powered or not? First, check the package for a “batteries included” or “batteries required” notice. Or, you can get a USB microphone, the best way to go. These are usually powered. They’re also Plug and Play, virtually guaranteeing success.
Once you have the microphone and the SRE software, you’re ready to get started. In this column, I’ll explain how to load the SRE from Microsoft Word in Office XP.
If you are first installing Office XP
|
1.
|
Choose the Custom installation type.
|
|
2.
|
In the Features to Install section, expand Office Shared Features and select Alternative User Input as seen below in Figure 1.
|
|
3.
|
Click Update, and the SRE will install.
|
If you previously installed Office XP, but still need to install the SRE component
|
1.
|
Put the Office XP CD-ROM into the drive.
|
|
2.
|
Click Start, click Control Panel, click Add/Remove Programs, select Microsoft Office XP, and click the Change button.
|
|
3.
|
Click Add or Remove Features and then click Next.
|
|
4.
|
This will reveal the installation options dialog show above in Figure 1.
|
|
5.
|
Expand Alternative User Input, click Speech, and then click Update.
|
After you have installed the SRE, you should open Microsoft Word and enable speech recognition.
To enable speech recognition in Microsoft Word
|
•
|
On the Tools menu, click Speech as shown in Figure 2 below.
|
When you do, you may or may not be prompted for the Office XP media again to finish the loading of the SRE. If you are, Word will load the rest of the SRE.
Once this is finished, you should close Word, and, even though not expressly directed to, you should reboot your machine. When you do, you’ll be ready to explore how Windows XP and the SRE integrate.
The steps to install the SRE from the Speech SDK are different than installing it from Office XP. In either case, however, the installation is relatively easy. The steps for training the SRE, configuring Windows XP, and using speech recognition as described below are the same, regardless of the source for your SRE software.
In addition to Microsoft Word, you can use speech recognition in almost any text field in any Windows XP program. You can control Windows Media Player, Internet Explorer 6, dictate into Notepad, or dictate e-mail into Outlook Express.
You’re now ready to show the SRE who’s boss. You do this via a little training of the SRE to make it recognize your personal vocal characteristics.
To train the SRE
|
•
|
Click Start, click Control Panel, click Sound, Speech and Audio Devices, and then click the Speech icon. The Speech Properties dialog box will open as shown below.
Figure 3
|
Each user of the SRE has his or her own unique profile. Chances are that you have different speech patterns than your sister. Windows XP and the SRE should take note of that fact. Theoretically, you can set up a different profile for everyone in the house, but, for this example, we’ll assume you’re the only user.
|
1.
|
Ensure that your microphone is working by clicking the Configure Microphone button.
|
|
2.
|
Click the Train Profile button to start the process. The Speech Recognition Training Wizard will open.
|
|
3.
|
Click Next, and follow the wizard instructions to train the SRE.
|
Now you’re ready to start using speech recognition via a tool in Windows XP called the Language Bar. At this point, you should configure the Language Bar, and other Windows XP speech settings.
To configure the Language Bar and Windows XP Speech Settings
|
1.
|
Open Control Panel.
|
|
2.
|
Click Date, Time, Language, and Regional Options, and then click Regional and Language Options.
|
|
3.
|
Click the Languages tab.
|
|
4.
|
Click the Details button.
Figure 4
|
The Text Services and Input Languages dialog opens, as seen above.
Clicking the Language Bar button allows you to configure how you want the Language Bar to appear. You can choose from four options such as having it float or having it docked at the toolbar.
You may change the recognition behavior of the SRE by clicking the Speech Recognition entry, then selecting Properties. For more about this, see the speech recognition documentation in Microsoft Word.
To enable the advanced speech recognition in Windows XP
|
•
|
Click the Advanced tab, which exposes how Windows XP can interface directly with the SRE as shown below.
|
|
•
|
Check the box beside Extend support of advanced text services to all programs.
|
|
•
|
Click OK, then, if required, restart your machine.
|
This option allows Windows XP to accept dictation-style input anywhere there is a box that can be filled in with text — including Notepad, Internet Explorer’s Address bar, and Outlook Express. However, this does not enable Windows XP to respond to commands for the operating system shell. You cannot use speech recognition to open the Windows XP Start menu, or the Windows XP Help and Support Center, for example. This is known as Voice Command mode, and while Office XP programs have this capability, Windows XP has it only in the Tablet PC Edition of Windows XP.
To test out your Speech Recognition capabilities, you should run two exercises. First, while running Word XP, locate the Language bar and click the Microphone icon as seen below.
Say “dictate” then say a sentence or two, and see that Word responds and enters in your text. In Word, you can also change the mode of input by saying “voice command.” This lets you navigate the menus by saying “menu,” or you can open Word Help by saying “help.” You can explore a lot of other intuitive voice-command features.
For your second exercise, start Notepad, and on the Language Bar, click the Microphone again and say “dictate.” You should be able to dictate into Notepad, which shows that Windows XP is interfacing with the SRE. You’re now ready to conquer any text-box and dictate your next e-mail message.
Now that solid speech recognition technology is available in Windows XP, I predict that we’ll see more third-party programs take advantage of the hooks that Windows XP provides. By setting up speech recognition now, training the SRE, and getting used to the interface, you’ll be ready to enjoy it when the next wave of integrated software arrives.
Expert Zone Columnist Jeremy Moskowitz is a Microsoft Certified Systems Engineer, and Microsoft Certified Systems Administrator, and the Chief Propeller-Head for Moskowitz, Inc. He is an independent consultant, and trainer for Windows technologies. He can be reached at jeremym-ms@Moskowitz-inc.com. Note that he cannot respond to all e-mail messages.
Haksa_Cloud 