May 30, 2019
My name is Ahmed, I’m 22 years old and I am studying for a master’s degree in Computer Science, This is my first year at GSOC, I am very passionate about NLP and Robotics. This year I will be working on A Natural Language Understanding (NLU) component for RoboComp
Significant efforts have been carried out in the Human Robot Interaction (MHRI) research field to provide multimodal and user-friendly interactive experiences between users and robots. As between humans, speech is the most natural, simplest and ideal modality to interact with robots using natural language processing technique, in particular when interacting with social robots.
The main purpose of this project is the study of different SRL systems freely available, mainly those based in Deep Learning techniques and will have to be adapted for their use in the RoboComp cognitive robotic architecture and Create a component capable of understanding the natural language of speech input sentences and provide API for this system, Moreover, these models are critical for automating tasks that require natural language understanding as part of a workflow.
My main goal is to build an open source software (end-to-end) and disassociate ourselves from Google Dialogflow which will support English, Arabic and Spanish languages that can be used for human-robot interaction with the aim of navigating in environments with people.
- TTS (text-to-speech, audio output, speech synthesis)
- STT (speech-to-text, audio input, speech recognition)
- Sentiment analysis