Child-Friendly Programming Interfaces to AI Cloud Services

AI cloud services are available for speech synthesis, speech recognition, image and video recognition, text analysis, and machine learning. School students could use these services in a wide variety of programming projects including voice commands to robots, chatbots, audio games, and vision-based robotics. In doing so they may learn about perception, language, psychology, and the latest empowering technologies. A major obstacle to using these services in schools is that they are technically complex APIs beyond the ability of most school students. The challenge addressed in this paper is how to provide interfaces that are much easier to use and yet still supports most of the functionality of these AI services. We describe the addition of new programming blocks to the Snap! visual programming language [1] that provide easy-to-use interfaces to these services. We have developed new blocks for speech input and output and image recognition. Learning materials have been developed and preliminarily trialed with a small number of children.