== Support for explicit prosody specifications ==
MARY TTS support explicit prosody specification using 'prosody' element as described in [http://www.w3.org/TR/speech-synthesis11/#edef_prosody W3C Speech Synthesis Markup Language (SSML) 1.1 recommendation]; and the different attributes in 'prosody' element like 'rate', 'pitch' and 'contour' are used as specifications to modify predicted phone durations and pitch contour before passing them to synthesizer.
'''Note:''' The following examples will only work if the `INPUT_TYPE` is set to `RAWMARYXML`. HSMM voices are able to realize the specified prosody out of the box; for unit selection voices, at least MARY 4.3.0 is needed, and the checkbox "Apply prosody modification" must be activated in the [http://mary.dfki.de:59125 web interface].
Example:
{{{
Welcome to the world of speech synthesis!
}}}
=== Adjusting Speech Rate ===
Example 1:
{{{
I am now speaking twice as fast.
}}}
Example 2:
{{{
I am now speaking 30% more slowly.
}}}
=== Adjusting Pitch level ===
Example 1:
{{{
I am now speaking with 3/4 pitch.
}}}
Example 2:
{{{
This is five semitones lower.
}}}
=== Shaping intonation contour ===
Example 1:
{{{
I am talking with rising intonation.
}}}
Example 2:
{{{
I am talking with falling intonation.
}}}