Remote Voice Control of Computer Based on Convolutional Neural Network

This paper dwells upon the solution of the task of creating the voice assistant to control the personal computer based on Windows in English using third-party language engines and the voice recognition models. The Python programming language, the Speech Recognition and PyTorch library are used to solve this problem. The article analyzes similar programs, tells about the stages of software development, demonstrates the program demo version example and analyzes the prospects for project work. The created model for speech recognition, which was developed during the implementation of this research has the following results as average loss 0.4305, average WER 0.3428 and average CER 0.121629.