GSoC’20 RoboComp project: An Automatic Speech Recognition Component for Robocomp

8th June 2020

About me

I am Anshul Lahoti. I am a fourth year undergraduate pursuing B.Tech and MS by research in Electronics and Communication Engineering from the International Institute of Information Technology, Hyderabad, India. I am pursuing my research in speech domain and currently doing my research work on Automatic Speech Recognition in a noisy environment. My interests lie in Speech Recognition, Medical Speech, Machine Learning, and Software Development. This year, I will be working on the Automatic Speech Recognition Component.

About the Project

Since the evolution of machines, there has been a need to make their interaction with humans user-friendly. A number of modes of communication have been considered for this purpose. Speech, being the natural mode of communication among humans, is a lucrative option to make the Human-Robot Interaction (HRI) as free as possible. Hence, in this project I will be using the current Speech Recognition technologies (deep speech techniques and libraries) that can be used in the RoboComp framework. The incorporation of speech recognition would mean that the spoken utterances of a person would be converted to text for the understanding of the robot. Automatic Speech Recongiton component develops in this project will be able to acheive following objectives-

1. It will transcribe speech to text with high accuracy.
2. It will provide Human-Robot Interaction.

ANSHUL LAHOTI