Difference between revisions of "Touchscreen"

From Engineered Arts Wiki
Jump to navigation Jump to search
(Introduction)
(Introduction)
Line 7: Line 7:
 
== Introduction ==
 
== Introduction ==
  
[[File:Socibot language screen.jpg|Right|400px|Language]]
+
[[File:Socibot language screen.jpg|right|400px|Language]]
  
 
This interface allows easy interaction with your robot.
 
This interface allows easy interaction with your robot.

Revision as of 13:35, 19 March 2016

This is a protected page.

Introduction

Language

This interface allows easy interaction with your robot.

The kiosk interface is accessible via a optional touchscreen if it was supplied with your robot, or you can access a browser based version remotely see here.

Kiosk Direct Browser Access

Note that all the following screens can be hidden if desired. You have control over what users will see. The following screens are usually enabled by default:

Language

The language page is the landing page, allowing you to select from the interface translations pre-installed on your robot.

Buttons

Quickly direct your users to the most important/relevant content using large, colourful buttons.

These can be used to play sequences or start control functions. Buttons can trigger a story, or a song, start SHORE age-gender-mood estimation, or play user created sequences from the Compose screen.

A 'stop' button halts any of the behaviours or sequences on this page.

Buttons (SociBot)

Library

The library screen allows for easy triggering of sequences and animations that are stored on the robot.

Simply press a button to trigger the desired animation. The arrow keys at the bottom of the screen move through further pages of animations.

Library (SociBot)

Custom content can be created by Engineered Arts, developed in conjunction with customers, to cover any subject desired.

You can also create and edit sequences yourself - see Understanding Sequences for information on how to do this.

Compose

The compose screen is used to quickly and simply generate new sequences.

Eye graphics, cheek colours, head movements and poses can be overlayed onto speech and audio effects to create a complete sequence. Sequences are saved as you go along and remain accessible until deleted.

Compose screen.jpg

  1. Audio, eye graphics, cheek colour, head movement and pose selections
  2. Timeline – drag selections here and rearrange as desired
  3. Start and Stop controls to play sequence on RoboThespian
  4. Timeline scrollbar – drag slider or press left and right arrows to scroll through sequence
  5. Sequence selector, currently 10 sequences can be created - this can be expanded. Press on any of the numbers to select the sequence to work on. Sequences should be saved when RoboThespian is shut down. NB please follow the shutdown procedure for RoboThespian - RoboThespian Shut Down
  6. Text-to-Speech Editor – click to create new speech or drag user created speech here to edit
  7. Bin – drag animations or audio here to remove them from the sequence. You can also drag any of the numbered buttons on the right (sequence selector) into the bin to delete the entire sequence.


Adding audio or speech

Existing audio and sound effects can be dragged with a finger to the time line in the bottom half of the screen or you can create your own speech using the text-to-speech engine.

Admin-compose-adding-audio.jpg

To see more selections, click on a button and sweep your finger to scroll through the available options.

Admin-compose-adding-more-selections.jpg

Adding own speech

You can create your own speech by clicking on the Create button, this opens an onscreen keyboard and text editor.

The Text to Speech engine can be pre-configured with a range of voices. To change the voice and language in use, click on the appropriate flag.

Compose own text.jpg

Type the desired text, test it with the play button, then hit the add button to add to your sequence. Should you later wish to edit your text, drag the text onto the Create / Edit box on the right.

Compose own text2.jpg

Adding animations

Eye graphics, cheek colours, head movements and poses can be overlayed onto the audio. Drag the desired selection onto the timeline to build your sequence.

Compose adding animations.jpg

Use the play and stop buttons to play the sequence on RoboThespian. Moving the scroll bar next to the play and stop buttons scrolls through the timeline.

Animation tips

A few tips on developing your own animations.

Some effective COMPOSE moves

Try and incorporate regular blinks, move the eyes around to look at your audience.

(If there is no blink function on your COMPOSE screen contact Engineered Arts to add it for you)

For example, blink, then look left with the eyes, turn the head slightly later to the left, make the eyes look straight again, look right with the eyes, then turn the head to the right, look straight with the eyes. Look forward again.

Try looping these gestures for expressive movement:

Effective COMPOSE moves 1.jpgEffective COMPOSE moves 2.jpg

General animation tips

  • Use sharp specific gestures
  • Listen to the script / audio
    • Think: where should key gestures be
    • Less is more - think how people gesture when they present information - they don't wave arms around all the time
    • Then fill in gaps if not moving enough - small / slow smooth movements

Modifying the Text-To-Speech voice (Acapela)

The Acapela Text-To-Speech can be modified with tags.

Use tags in the text entry field.

For example:

 \rspd=120\My name is Robo Thespian. I can spell that\rms=1\robothespian.\rspd=70\I can talk slower if you like.
 \vct\=75\I am angry!

Live

The live screen allows for live control of the robot. You can make the robot look in any direction, by touching the screen section showing the view from its head webcam. You can also get the robot to say different greetings, farewells, questions, responses, and even make noises; move the head in pre-defined sequences, or change facial expressions or eye positions.

Live (SociBot)


Clockwise from upper left

  1. Pupil control - move pupils to point where you like
  2. Eyelid control - blink, wink, shut eyes, open eyes
  3. Novelty pupils - replace Socibots irises with a dollar sign, heart, spiral, or back to the default (blue eyes)
  4. Cheek colours - select colour and brightness/intensity from the colour picker and brightness slider
  5. Head tilt slider - press on the slider and drag to tilt the head sideways
  6. Audio / speech controls - press on desired text or the speaker button next to the text to make SociBot speak or make sounds
  7. Audio / speech bank selector - use to select different categories of audio for SociBot - there are several menus
  8. Mood selection - choose from several default expressions: happy, sad or surprised
  9. Head movements - move left, right, up, down, shake and tilt the head
  10. Quick face selection - black woman, white woman, bearded man, Queen, Ziggy Stardust, skull
  11. Live video feed - from the head mounted camera. Use this to control the robot's head and gaze direction manually.


Sensors

When the sensors screen is displayed, the robot will engage background behaviours such as looking at new people, or tracking the hands of users. The sensors screen displays camera feeds from the chest-mounded RGB and depth sensor, and head camera with relevant overlays (hand being tracked, user's joint position, (optional) SHORE age-gender-expression data.

The depth sensor is used by SociBot to detect people, track and focus on an individual it is interacting with, and detect the body positions of people in its field of view.

At the top of the screen there are two views:

  • The left view shows the image from the chest-mounted colour camera
  • The right view shows the image from the infra red camera, used to help calculate depth or distance to objects in its field of view


Sensors (SociBot)


Administrative Screens

If you have administrative access to the robot, you can use the admin touch-screens to debug, control and even manage the content on the robot. The touchscreen CMS has been superseded by the browser-based CMS, however older Socibot models may still have this functionality.

See Kiosk Admin for a detailed guide.