VoiceXML Tag <say-as> Version 2.0

GetVocal Voice Portal Technologies

[1.0 to 2.0][Tags][Attributes][Variables][Events][Properties]

VoiceXML Tag
<say-as>

Speak value.

Parent Tags
<audio> <choice> <emphasis> <enumerate> <p> <paragraph> <prompt> <prosody> <s> <sentence> <voice>

Content Syntax
The value must be parsed text.

Attributes

Name	Req/opt	Default	Valid values
sub	Optional		Text
type	Required		"acronym" "address" "currency" "date:d" "date:dmy" "date:m" "date:md" "date:mdy" "date:my" "date:y" "date:ym" "date:ymd" "duration:h" "duration:hm" "duration:hms" "duration:m" "duration:ms" "duration:s" "measure" "name" "net" "number:digits" "number:ordinal" "telephone" "time:h" "time:hm" "time:hms"

Compatibility
VoiceXML 1.0 No, 2.0 Yes.

Description
The <say-as> tag specifies how to interpret text for speech synthesis.

The text is either specified as the value of the tag (i.e., the text between the start and end tag), or by the sub attribute. If the sub attribute is specified, the value of the tag is ignored.

The type attribute specifies the data type of the text to be spoken. Note that type="acronym", perhaps counter-intuitively, means to speak a group of letters as a word, as in DEC or NASA, rather than the default which is to spell out the letters, as in IBM or USA.