This documentation is archived and is not being maintained.

Collecting Confidential Data

Speech Server 2007

This content is no longer actively maintained. It is provided as is, for anyone who may still be using these technologies, with no warranties or claims of accuracy with regard to the most recent product version or service release.

Application developers should consider the sensitivity of the information that the application collects and uses. The results of speech recognition, the text of prompts, audio input and output, and the contents of resources (such as grammars) are all logged in the Windows event tracing log (ETL) files unless flagged as confidential. When data is flagged as confidential, it is not recorded in the logs.

Flagging data as confidential varies depending on the programming model for the application.

For a managed code application, use LogResults property of various speech recognition, speech recording, and speech synthesis objects. By setting this property to false, results from the activity are not be logged. For more information, see the following properties:

For a VoiceXML application, you control the confidentiality of recognition results, key presses, and prompt audio using a proprietary setting of the property element (standard VoiceXML does not support this functionality). For more information, see Design VoiceXML Applications for Easy Reporting and Tuning.

For a SALT voice response application, use the confidential parameter of the prompt, listen, and dtmf elements. For more information, see param Element.