Supports the insertion of recorded audio files.
<ssml:audio src="string" /> <ssml:audio src="string"> </ssml:audio>
| Term | Definition |
|---|---|
|
src |
Required. Specifies the name of the audio file. The value of this attribute can be an absolute, local path for speech applications running in a local environment, but it must be an HTTP URI for applications running in a Speech Server environment. This attribute is case-sensitive only when referencing files residing in file systems that are case-sensitive. |
The ssml:audio element supports RIFF header .wav files encoded with the following parameters.
| Parameter | Value |
|---|---|
|
Format |
PCM, a-law, u-law |
|
Bit-depth |
8 bits, 16 bits |
|
Channels |
Mono only |
|
Sampling Rate |
All sampling rates supported. |
Use this element to play a known audio file.
For speech applications deployed in a Speech Server environment, headerless audio files can be used provided that the server supplying the files passes the MIME type of the files.
The content of a non-empty ssml:audio element should be the plain text or SSML marked-up text to be spoken if the audio file specified in the src attribute is either not available or unplayable. The content of a non-empty element can also include another ssml:audio element.
<?xml version="1.0" encoding="ISO-8859-1"?> <ssml:speak version="1.0" xmlns:ssml="http://www.w3.org/2001/10/synthesis" xml:lang="en-US"> <ssml:p> Thank you for volunteering your opinion. <ssml:audio src="http://www.treyresearch.net/audioprompts/opinionprompt.wav"> Please voice your opinion after the beep. </ssml:audio> <ssml:audio src="http://www.treyresearch.net/audioprompts/beep.wav" /> </ssml:p> </ssml:speak>