Speech-enable your browser app

Speech-enabled elements

By default, the following elements on your web page are speech-enabled:

• <input> controls with type="text" (the default value).
• <textarea> controls.
• <div> elements with the HTML5 contenteditable attribute.

Procedure

To implement core speech recognition functionality when your page loads, do the following:

1. Configure the NUSA_Guids cookie with your organization token and partner GUID.

2. Link your page to the Dragon Medical SpeechKit (Browser edition) JavaScript file.

3. Provide the current user name and app name.

You can also load your page without speech recognition and initialize speech recognition at a later point from your integration. For more information, see Deferred initialization.

Configuring the NUSA_Guids cookie

Set the NUSA_Guids cookie from your web server according to the partner GUID and organization token you receive with your license information:

NUSA_Guids = <organization token>/<partner GUID>

It's also possible, but not recommended, to set the NUSA_Guids cookie using JavaScript. To do this, add the following to your NUSA_configure() function or before you call the NUSA_initialize() function:

document.cookie="NUSA_Guids=<organization token>/<partner GUID>";

Linking your page to the Dragon Medical SpeechKit (Browser edition) JavaScript file

For Nuance-hosted systems, Dragon Medical SpeechKit (Browser edition) functionality is provided by Nuance.SpeechAnywhere JavaScript files specific to your region and release channel. An alternative set of files is provided to implement deferred correction. Use the URL specific to your region and release channel. For a list of URLs, see: Release channels.

Add the JavaScript file to your page header; for example:

<script type="text/javascript" src="https://speechanywhere.nuancehdp.com/mainline/scripts/Nuance.SpeechAnywhere.js"></script>

Important: Make sure the Nuance.SpeechAnywhere JavaScript file is added to your page header; it shouldn't be added to a different place in the page.

If you're deploying Dragon Medical SpeechKit on-premises at your own site, point to the Nuance.SpeechAnywhere.js file on your web server. For more information, see the Dragon Medical Server Installation Guide delivered with your on-premises package.

Providing the user name and app name

Copy the following into your page header and set the parameters:

<script type="text/javascript">
    function NUSA_configure()
    {
        NUSA_userId = "<user name>";
        NUSA_applicationName = "<your application name>";
    }
</script>

When your page is loaded, Dragon Medical SpeechKit calls your NUSA_configure() function and speech recognition is enabled for the page. Make sure to do this for every page that you want to be speech-enabled in your app.

See also

Deferred initialization of Dragon Medical SpeechKit

Using HTML5 content-editable elements

Speech-enabling only specific controls

Starting and stopping recording

Naming fields for voice navigation

Styling speech-enabled controls

Positioning the speech bar