Share via

RecordSound Remarks

  • Article
  Microsoft Speech Technologies Homepage

The RecordSound control is provided to record user speech, and to copy the recording from the speech server to the Web server for playback. The RecordSound control can be used only on the Telephony Application Services server and within Telephony Application Simulator. It cannot be used on other clients.

The folder on the Web server in which recordings are saved must be shared in such a way that its contents can be accessed by the Speech Server. The RecordSound control's SavePath property should contain the full local path of this folder on the Web server, and the PlaybackUrlBase property should contain the domain and the name of this folder.

The RecordSound control contains an internal QA control. The RecordSound control delegates the playing of its prompt to the Prompt element of this QA control, and it delegates the collection of the user response to the Reco element of this QA control. The RecordSound control uses its InitialTimeout, BabbleTimeout, MaxTimeout, and EndSilence properties for the purpose of controlling the internal Reco element.

The RecordSound control saves information about the record file in its SemanticItem.

StopOnDtmf Property

While a speech application runs, the dual-tone multifrequency (DTMF) inputs the caller enters are stored in a DTMF buffer. Typically, when a speech control with a DTMF grammar is activated, the contents of the DTMF buffer are compared with the DTMF grammar and then cleared. However, the RecordSound control uses the DTMF buffer in a way that is different from the other speech controls. When the RecordSound control is activated, it does not clear the DTMF buffer. Also, note that the StopOnDtmf property of the RecordSound control is often used to stop recording. If this property is set it will be included in the buffer, and will interfere with the input for the next DTMF control.

It is useful to keep all of this in mind when considering how to handle application logic for successive speech controls that handle DTMF input. The solution is to either set the DTMF PreFlush to true on the next control, or provide some other mechanism to account for the extra input in the DTMF buffer.

Contents

The RecordSound control can contain the following controls:

  • Prompt
    A DialogPrompt control that plays a message to the user.
  • Params
    An optional Params collection, containing a Param control that configures the Prompt.

Markup

<speech:RecordSound properties runat="server">
  
  <Prompt properties ></Prompt>
    
  <Params>
    <speech:Param Name="...">...</speech:Param>
  </Params>
    
</speech:RecordSound>

See Also

RecordSound Class | RecordSound Constructor | RecordSound Members | RecordSound Properties | RecordSound Methods | RecordSound Events