Simulation and Implementation of English Speech Recognition by NLP
K. Kavita, K. Suresh Kumar, Sridevi Dasam, Kiran Sree Pokkuluri · 2025
The use of hybrid deep learning-based voice recognition in oral English practice, in conjunction with multimodal natural language processing education, begins with an introduction to the fundamentals of speech recognition technology. An explanation of the hidden Markov model and its three essential algorithms is provided, followed by the realization of its simulation and use in voice recognition. The system's architecture and essential technologies are presented. First, the text delves into the use of deep learning in natural language processing and recording oral English instruction by specialized instructors. Each person has their own optimal reading time and preferred phrases. In all, there are several individuals. The phrases used are spoken English, so it would be advantageous to provide a course in spoken English to help people enhance their oral communication skills. The findings of the trial indicate a decrease in the accuracy of identification, but a tenfold increase in recognition speed. Another advantage is that the scoring system is equally accurate to the platform system. By validating the feasibility and effectiveness of this approach, it enhances the accuracy of instruction categorization. Attention mechanisms will be utilized to expand this strategy in the future.