Speech Recognition System for Upper-Limb Impairment: A Text Editor for Programming via Voice

Programming is one of the most important skills to learn for the present and future generations. However, those with hand disabilities may find it difficult to program with a traditional keyboard and mouse. Disabled people would greatly benefit from assistive technology that allows them to use their voices to control computers. The existing system recognizes user input as voice and translates it into a program. However, the existing system only recognizes a limited number of keywords in a programming language. It would also be difficult to debug errors if the voice was misinterpreted. Our suggested work in speech recognition using natural language processing (NLP) would perform better than the previously available systems. It is possible to build a foundation for programming without a keyboard by creating tools that use a publicly accessible speech recognition APL Voice control of the user interface for program execution and code debugging would also be quite beneficial. This can be done with the help of modules such as Pynput. Keyboard, PyAutoGUI, which can simulate mouse cursor movements and clicks as well as keyboard button presses. The accuracy of our proposed system is 87%. When converting voice commands to text, the Word Error Rate (WER) is 18%.