GITHUB directory: https://github.com/francesco-AI/AI-SHRI-Project
For my Spoken Human-Robot Interaction project I chose to implement a robot gardener capable of performing a series of actions in the field of plant care and maintenance. The chosen context allowed me to implement a series of dialogues through which to experiment with the use of speech-recognition and text-to-speech routines: through the "Google Speech Recognition" and "SAY" libraries it was possible to send and receive voice commands in my source language (that is to say: the Italian). But in particular I stopped to study the StanfordNLP library for the treatment of the recognized text. As for the semantic part, I studied the theoretical concepts that underlie the FRAME theory. I only used the FRAMENET library at the beginning of my project, when I still hadn't decided to implement everything in Italian. Unfortunately Italian is not one of the languages supported by the current library and therefore, even having used it, it was not possible to insert it in a "thorough" way in this project, even if I took it into account at least as an approach in the recognition phase of the actions to accomplish.
