Games for the summer 2019 Tommy Thumb study - both are in the folder "Final Versions"
This game is part of the Tommy Thumb system. It is meant to be played on the QTRobot built by LuxAI running ROS Kinetic on Ubuntu 16.04. The game is run during an experimental set up combined with several other subsystems (each on their own computer/microprocessor except CoRDial, which is run on QT's body computer): orthosis, camera, and EMG.
Before running the game a package must be made for the games and the catkin workspace set up. Several functions in the game use ROS publishers developed on QT or through the other subsystems and these must be set up as well (see links for QTRobot wiki, orthosis, camera and CoRDial).
Git pull from the repository linked above. In the first terminal window run:
> roslaunch cordial_example test_setup.launch
In the second terminal window run:
> roscd cordial_face/web
> http-server
Look for the number of the available ports - use this port number (it is 8081 in my case) for the third window (first switch to QT's wifi and ssh to QT's head computer):
> ssh qtrobot@qtrobot
> luakit -U 192.168.100.2:8081/KiwiLite.html
This command runs the web browser in private mode using the body computer's processor, the available port and the KiwiLite version of CoRDial. The CoRDial face should show up on the screen after this command is run.
In the fourth window run your python game script (you can run this on any wifi and it should be on the body computer):
> rosrun packagename QTNumberGame.py
In order to add new things for QT to say, follow these steps. First, find the script file where all text statements are kept. On the QT104 body computer this is located in ~/catkin_ws/src/cordial-public/cordial_example/speech/script.txt. Don't push this to the git repo for cordial if it is modified, make a new file and package instead. Every line in script.txt should follow this format:
[statement6]I can also perform with amazon effects!
[statement7]*loud*You can <prosody volume="x-loud">make my voice loud</prosody>, *quiet*or <amazon:effect name="whispered">soft as a whisper</amazon:effect>.
[statement8]<prosody rate="fast">I can speak fast when I'm excited</prosody>, <prosody rate="x-slow">or slow down to explain things</prosody>.
The file ~/catkin_ws/src/cordial-public/cordial_example/CoRDial_behaviors.json has timing and names of all the emotions/behaviors that CoRDial can make - these are included between ** in the statements. Any spoken text is written out. Do not put extra line breaks in the file or it will stop speaking at the break.
In the .../speech/ folder send the text through AWS to generate audio files. Do this by running this from the command line:
> ./gen_audio.sh
Next we need to send the audio files to the head computer. Connect to QT's wifi, then run:
> ./send_to_qt.sh
Now the speech files are available for playback in the game.
https://docs.aws.amazon.com/polly/latest/dg/supportedtags.html