The Speech Synthesizer Message Set defines the Hub message for text-to-speech usage.  The message set consists of messages to send to a server that contains a text-to-speech engine.

The Hub uses speech synthesis engines that support  SAPI and JSAPI message as defined in the speech synthesis standards.  Message wrappers translate the Hub messages into JASPER and SAPI messages.  The speech synthesis JSAPI implementation describes the features of the speech synthesizer could be supported by a JSAPI implementation using IBM Via Voice.

If other speech synthesis message sets need to be supported by the Hub, please send a message to the MITRE Communicator team at bugs-darpacomm@linus.mitre.org.

This Speech Synthesis Message set is part of the Speech Message Set.

• Usage
• Message Set
• Messages Issued

Usage
 
 

Command Line
 
 

Default Port

15570

Command Line Arguments

None

Message Set

• reinitialize
• Configure
• Get_Word_Pronunciation
• GetVoice
• Pause
• Resume
• Pronunciation
• Remove_Word
• Set_Voice
• Speak

reinitialize: sets up the connection to the speech synthesizer.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:audio_destination_type	GAL_STRING			Describes where the output is going to:  <"brokering"| "file" | "speaker" | "telephone" >
	audio_destination_param	GAL_STRING		audio_destination_type	The name of file or port number (for brokering, speaker, or telephone) to use.
	:average_speaking_speed	GAL_INT	yes		Sets the average words per minute for the synthesis engine.
	:phoneme_translation_file	GAL_STRING	yes		Name of the file that contains the grammar for the synthesizer to use.  For example, in SAPI the contains are word phonetic translation table, while in JSAPI file contents is the grammar written the Java Speech Markup Language (JSML).
	:pitch	GAL_FLOAT	yes		Sets the average speaking rate for the synthesis engine.
	:real_time_setting	GAL_FLOAT	yes		A value that indicates how close to real time the synthesis process should take, where a value of 1.0 indicates real time.  Value is between zero and one.
	:volume	GAL_FLOAT	yes		Sets the volume level for the recognizer.  Value is between zero and one.

Configure:  This message defines that type of type of messages to expect from the synthesis server.  By enabling one of these values, the synthesizer issues issue a corresponding message.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:current_byte	boolean	yes		A value of true causes the synthesizer to issue the  Current_Byte message to indicate the current byte of the audio stream the synthesizer is processing.
	:marker	GAL_STRING	yes	:notify_status	Defines the symbol for the synthesizer to issue the Marker_Reached message which indicates where the synthesizer is in the generation of the spoken utterance.   Note that that the :notify_status must be true to receive the Marker_Reached message notifications.
	:notify_attribute_changes	boolean	yes		A value of true causes the synthesizer to issue the Attribute_Changed message when an internal attribute value changes.
	:notify_status	boolean	yes		A value of true causes the synthesizer to send Synthesize Event Message indicating different synthesis events.  (Default value is true.)
	:notify_word_position	boolean	yes		A value of true causes the synthesis to issue the Word_Position message that indicates the current word position of the audio stream being synthesized.

Get_Voice:  This message causes the synthesis process to indicate the current voice models.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:default_voice	GAL_STRING			The name of the default voice model for the synthesize to use.
	:voice_tags	GAL_LIST	*		A list of tags that the synthesizes uses to generate different voices.
	:voice_names	GAL_LIST	*	:voice_tags	A list of names that corresponds to each voice_tag entry.

Get_Word_Pronunciation:  This causes the synthesizer to return the phonetic string of a given word.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:word	GAL_STRING			The word to determine a given  phonetic sequence.
OUT:	:phone_sequence	GAL_STRING			The phone sequence of the given word.

Pause:  This message pauses the synthesis process until the Resume message is issued.  It is possible that some text message will be lost when the synthesizer  is in a paused state.
 

	Parameter	Type	Optional	Depends on	Description / Constraints

Resume:  This message resumes a synthesizer that is currently paused (see the Pause message).
 

	Parameter	Type	Optional	Depends on	Description / Constraints

Pronunciation:  This message defines how the synthesis is to define the phone string to pronounce certain words.  This message can be used to modify the pronunciation of an existing word or to add a new word.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:words	GAL_LIST			A list of words.
	:phones	GAL_LIST			A list of phonetic strings to use to pronounce the corresponding word.

Remove_Word:  The message removes the word and its pronunciation from the synthesis process.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:word	GAL_STRING			The word (and associated) pronunciation to remove.

Set_Voice:  The message specifies the voice model for the synthesizer to use.
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:voice_default_name	GAL_STRING			A string that identifies the default voice model to  use.
	:voice_tags	GAL_LIST	*		An list of voice tags for synthesizer to use.
	:voice_names	GAL_LIST	*	:voice_tags	A list of voice names that corresponds to each voice_tags.

Speak:  The message causes the synthesize to output the given text.  (The text may contain marker characters to receive notifications.)
 

	Parameter	Type	Optional	Depends on	Description / Constraints
IN:	:text	GAL_STRING			The message to synthesize.

Messages Issued
 
 

The synthesizer issues the following messages to the Hub:.

• Attribute_Changed
• Current_Byte
• Marker_Reached
• Synthesis_Event
• Word_Position

Attribute_Changed:  The synthesis process  issues this message when a request has been made using the Configure message to receive notification of when certain attributes change their value.  The actual name of the attribute depends on the implementation of the synthesis process.

{c Attribute_Changed ... }
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:attribute_name	GAL_STRING			The name of attribute.
	:attribute_value_new	GAL_STRING			The value of the attribute.
	:attribute_value_old	GAL_STRING			The previous value of the attribute.

Current_Byte:  The synthesis  process  issues this message when a request has been made using the Configure message to receive notification the current byte of audio stream being processed.  The frequency of this message depends on the synthesizer.

{c Current_Byte ... }
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:current_byte	GAL_INT			The current audio byte being processed within the synthesizer.

Marker_Reached:  The synthesis  process  issues this message when a request has been made using the Configure message to receive notification the when a certain marker appears in the text stream.

{c Marker_Reached ... }
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:marker	GAL_STRING			The marker found in the text string.

Synthesis_Event:  The synthesis process issues this message when an event unique to the synthesizer after the notify_status parameter has been enabled by the Configure message.  The actual parameter of the event_name depends on the synthesizer.  For example, a SAPI synthesizer sends events for reaching  markers (bookmarks), audio begin, and audio end.  A JSAPI recognizer generates similar events for encountering a marker in the text and for beginning and ending of the text-to-speech generation.

{c Synthesis_Event ... }
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:event_name	GAL_STRING			The value describes the type of event and depends on the synthesizer being used.

Word_Position:  The synthesis process issues this message when a request has been made using the Configure message to receive notification of the current word position in the text-to-speech generation process.

{c Word_Position ... }
 

	Parameter	Type	Optional	Depends on	Description / Constraints
OUT:	:position	GAL_INT			The word position of the current text-to-speech utterance.