Voice Mode

The voice mode attribute sets the current voice mode. Within SGX, "voice modes" refer to different types of vocalization—such as effort sounds, breaths, or laughter—which can be reflected in the animation. The default value of this attribute is called "default", which indicates normal speech.

Currently, voice modes are realized as moods in the Mood Library. Using a non-default voice mode activates the corresponding mood. Using the default voice mode has no effect on animation.

Automatic Voice Mode

Automatic Voice Mode is a feature in SGX that automatically detects voice modes and activates the associated expressions. At present, one voice mode, "effort", can be automatically detected, so it is useful to set up an "effort" mood in the Mood Library.

Software Requirements

Using the Automatic Voice Mode feature requires SGX 3.4 or newer.

Resource Requirements

Automatic Voice Mode requires you to have the following mood in your Character Definition:

effort - This mood should contain a variety of expressions that convey tense exertion on the face. This voice mode will be triggered when physical effort sounds like grunts are detected.

Note: The name of the voice mode in your Mood Library must exactly match the above in spelling and case in order to be triggered.

Setting Voice Mode

Set Voice Mode for a Batch (via SGX-GUI)

SGX-GUI allows you to set voice mode for an entire batch through the "Voice Mode" field in the "Batch Markup" portion of the UI. To activate Automatic Voice Mode, enter "auto" into the Voice Mode field. Entering "default" prevents the application of any Voice Modes. Entering a specific voice mode such as "effort" in this field would cause the entire batch to exhibit that voice mode. Auto is usually the best option if the detectable voice modes occur in the batch.

Set Voice Mode for a Batch (via Command Line)

To set voice mode for an entire batch on the command line use the markup option -m; for example, -m "voice_mode=auto".

Set Voice Mode(s) in the Transcript

Multiple voice modes can be applied manually to a single line of audio by inserting markup tags in the transcript. The transcript voice modes override the default voice mode or any voice mode set for the entire batch. Each markup tag containing a 'voice_mode' attribute applies to everything that follows it up until the next tag containing a 'voice_mode' attribute. For example:

CODE

I can't reach the handle (voice_mode=effort) it's too high
(voice_mode=default) can you reach it for me

In this example:

"I can't reach the handle" will have default voice mode or the voice mode set for the entire batch.
"it's too high" will have voice mode effort.
"can you reach it for me" will have default voice mode.