I have always wanted to experiment with robotics, and lately I’ve found the time to build an “intelligent”, open-source robotic platform.
The platform runs embedded debian linux, and includes the following main capabilities:
1. Computer Vision (imitation of the human vision. The robot sees and “understands” what it sees). For this I have extensively used the OpenCV project.
2. Speech Synthesis (imitation of human speech. The ability to speak). For this I have used Espeak.
3. Speech Recognition (the ability to understand vocal commands). For this I have used CMU Sphinx 4 (after modifying some of the files in the project)
Just a teaser, before I upload more videos:
List of Computer Vision capabilities:
1. Line Following
2. Object Tracking
3. Facial Recognition (uses face API from face.com)
List of Speech Synthesis Capabilities:
1. Speaking the current action (“Moving Forward” etc)
2. Saying “Hi <name>” when it recognizes a face
Current list of Speech Recognition Capabilities:
1. Full movement control (“Move Forward\Backward”, “Turn Left\Right”, “Stop”)
2. Initialization of algorithms: “Follow Line\RedBall”, “Find Face”