Skip to main content

chatbotData

The parameter chatbotData allows you to set the size of the chatbot's video animations

NameRequiredTypeDefault valueDescription
sizeNonumber | string | undefinedDefines the height of the animations. Can be a string representing percentage (ex: "80%"), a proportion of view's height (ex: "50vh") or a fixed number of pixels (ex: 1500)
heightNonumber | string | undefinedMargin above the top of the animations (i.e. margin-top). Same values as size
fullWidthInDesktopModeNoboolean-true for the character's container to use full available width (centered character)
cameraPresetNostring of preset'default'(since v4.0.3, 3D agent only) Determinates the position of the camera that influences the character's display
calibrationNoCalibrationData-(since v4.0.3, 3D agent only) Modification of the height of the character's gaze when it's listening or speaking

preset​

"default" // character from head to knees
"head" // zoom on the head
"neck" // head and neck
"shoulders" // head, neck and shoulders
"chest" // from head to breast
"belly" // cut at the lower part of the belly (quite like a medium-shot)
"knees" // (like "default")
"feet" // long-shot

calibrationdata​

The gaze height calibration data allows you to modify the spatial coordinates of the point at which the agent’s gaze will be fixed when the agent enters active listening mode (the user performs voice recognition) or speech mode. There are two available modes, static and dynamic, which are mutually exclusive. Static mode allows you to set a height toward which the agent’s gaze will tend; this height remains fixed once entered. Dynamic mode has not yet been extensively tested in real-world situations, and the parameters required for its use must be entered precisely to ensure proper operation. Using facial recognition via a webcam, it allows the agent’s line of sight to adjust based on the height of the person standing in front of them. Calibration data is stored in local storage once entered, or can be entered manually during configuration. If no data is present, a calibration phase will appear when the application is launched.

NameRequiredTypeDefault valueDescription
staticNostaticData-Static mode
dynamicNodynamicData-Dynamic mode

static​

NameRequiredTypeDefault valueDescription
enabledYesbooleanfalsetrue to activate static mode
loadStaticTargetFromLocalStorageNobooleanfalsetrue to attempt to retrieve the data stored in local storage during initialization in order to bypass the calibration phase
staticTargetNo{ x: number, y: number, z: number }-Coordinates of a point in space that will be the character's line of sight. Note that the space generated by Three.js uses metric units, but each space differs depending on the 3D character used. It is recommended to use this option only when coordinates have already been obtained during a previous calibration on the same character

dynamic​

NameRequiredTypeDefault valueDescription
enabledYesbooleanfalsetrue to activate dynamic mode
loadStaticTargetFromLocalStorageNobooleanfalsetrue to attempt to retrieve the data stored in local storage during initialization in order to bypass the calibration phase
dynamicDataNodynamicData-Data used in the dynamic system. It is recommended to use this option only when data have already been obtained during a previous calibration on the same character

dynamicdata​

NameRequiredTypeDefault valueDescription
webcamHeightYesnumber-Height of the webcam (in cm)
ratioPixelToCentimeterYesnumber-Correlation between a pixel in the webcam rendering and a size in cm
deltaAgentHeightAgentWorldHeightInCentimetersYesnumber-Difference between the height of the top of the agent's skull on the marker and that in the 3D scene generated by Three.js (in cm)
cameraRotationInDegreesYesnumber-Camera rotation (in degrees). Sometimes webcams are positioned in such a way that you need to rotate the image 90° to get it oriented correctly.
distanceFromAgentToCameraInCentimetersYesnumber-Distance between the webcam and the user's location when using a kiosk (in cm)