-
Notifications
You must be signed in to change notification settings - Fork 15
Interruptions
We distinguished 3 types of reactions to an interruption:
- HALT: the current speaker halts and yields the turn to the interrupter.
- OVERLAP: the current speaker continues speaking and grabs the turn.
- REPLAN: the current speaker halts but immediately responds with a new utterance thus keeping the turn.
We noted for each type of reaction a variety of nonverbal behaviours that was common to all three types. Interruptees react with their eyes,gaze, head, smile, gesture and posture behaviours. More specifically, while reacting to an interruption they close the eyes, raised/lowered the eyebrows, gazed at the interrupter, tilted, nodded or tossed their head, smiled, held, retracted or expanded their current gesture for a variable amount of time, leaned forward their torso.Those behaviours are sometimes exhibited in concert with others, other times in isolation.
In the HALT case, the interruptee has two distinct gesture behaviours that we named GESTURE HOLD and RETRACT. Those correspond to two phases during the reaction where they held their current gesture and, sometimes, they retract towards a restpose with hands down by performing another pause (i.e. gesture freeze) of a certain duration that we called GESTURE RETRACT. In the OVERLAP case, the interruptee react by expantheding the current gesture, meaning that while overlapping their speech they enlarge the movement of their hands thus resulting in an expansion of their current gesture.
The objective of this toolbox, is to have a testing module that allows us to generate and observe synthetic nonverbal reactions
Connection with GRETA modules
Advanced
- Generating New Facial expressions
- Generating New Gestures
- Generating new Hand configurations
- Torso Editor Interface
- Creating an Instance for Interaction
- Create a new virtual character
- Creating a Greta Module in Java
- Modular Application
- Basic Configuration
- Signal
- Feedbacks
- From text to FML
- Expressivity Parameters
- Text-to-speech, TTS
-
AUs from external sources
-
Large language model (LLM)
-
Automatic speech recognition (ASR)
-
Extentions
-
Integration examples
Nothing to show here