All supported TTS engines can alter default text synthesis behaviour through the use of control tags, embedded tags, or escape sequences.
EXAMPLES
You can instruct the TTS engine to read a sentence quickly with excitement, or with a monotone voice (i.e. you can modify its prosody).
You can also specify how text should be analyzed. For instance, in some contexts a list of numbers such as 8390033 should be read as a telephone number; in others, it should be read as an actual value.
Although Mrcp defaults to use plain text, it also supports W3C's Speech Synthesis Markup Language, which is a sophisticated way of doing prosody. An example of how to do this follows.
<?xml version="1.0"?>
<speak>
<paragraph>
<sentence>You have 4 new messages.</sentence>
<sentence>The first is from <say-as type="name">Stephanie
Williams</say-as>and arrived at <break/><say-as type="time">3:45pm</say-as>.</sentence>
<sentence>The subject is <prosody rate="-20%">ski
trip</prosody></sentence>
</paragraph>
</speak>
For more information on the syntax used to support prosody text in Mrcp engines, refer to the SSML W3C recommendations at http://www.w3.org/TR/speech-synthesis/.
You should also check your Mrcp engine documentations for the subset of the SSML tags it is supporting.
Nuance's Speechify uses proprietary embedded tags to change the default text processing behaviour. You can use them to fine-tune pronunciations of words, phrases or sentences.
See the Speechify documentation on embedded tags for more details.