Speaker Directives

• EMPH (container element): set the emphasis of the contained text.
  • LEVEL (numeric, descriptive)
• BREAK (empty element): sets an intrasentential, prosodic break at current position.
  • LEVEL (numeric, descriptive)
  • MSEC (numeric)
  • TYPE (descriptive): a punctuation symbol that represents (roughly) the kind of intonation contour to be associated with the material preceding the break (e.g. `?' to mark ``question'' intonation).
• PITCH (container element): sets properties associated with pitch of the enclosed region.
  • BASE (numeric, descriptive)
  • MIDDLE (numeric, descriptive)
  • RANGE (numeric, descriptive)
• RATE (container element): sets the average speech rate of the enclosed region.
  • SPEED (numeric, descriptive)
• VOLUME (container element): sets the amplitude of the enclosed region in terms of the available range of the engine.
  • LEVEL (numeric, descriptive)
• AUDIO (empty): load and play an audio URL starting at the given point.
  • SRC: URL of audio document
  • MODE: specifies whether to play in background or not
  • LEVEL: level of audio relative to surrounding speech
• PRON (container): substitute the specified pronunciation for what would normally correspond to the contained text.
  • IPA: character string in Unicode IPA (International Phonetic Alphabet)
  • SUB: attempt at ``phonetic'' spelling in the language of the enclosing text
  • ORIGIN: ISO639 identifier for the language of origin of the enclosed text 
• LANGUAGE (container): specifies the language of the contained text.
  • ID: ISO639 identifier for the language
• SPEAKER (container): defines properties of the speaker speaking the contained text
  • GENDER
  • AGE (descriptive)
  • NAME: ``name'' of a speaker if a particular engine is being used

As a sample of the use of some of these tags, consider the following example from a hypothetical e-mail reader that uses SABLE markup. Since e-mail readers have access to information about at least some structural aspects of the input -- e.g. header information about the sender, subject and date, this information can be used to control the synthesizer's behavior in useful ways.:

<DIV TYPE="paragraph">New e-mail from
<EMPH>Tom Jones</EMPH>
regarding <PITCH BASE="high" RANGE="large">
<RATE SPEED="-20%">latest album</RATE>
</PITCH>.</DIV>
<AUDIO SRC="beep.aiff"/>

The subject information (``latest album'') is highlighted auditorily by setting a higher base pitch and larger pitch range, and by slowing down the speech by 20%. Finally, the header is terminated by an audible beep (``beep.aiff'').