AI-Powered Mock Interview System with Multi-Accent Support

Abstract

The increasing competition in today’s job market has made interview preparation an essential skill for students and job seekers. Traditional mock interview methods often depend on human trainers, which can be timeconsuming, costly, and inconsistent. To address these challenges, this paper presents an AI-Powered Mock Interview System with Multi-Accent Support that provides an automated platform for practicing interview questions and receiving instant feedback. The system allows candidates to respond using either voice or text input. Spoken responses are converted into text using Automatic Speech Recognition (ASR), while Natural Language Processing (NLP) techniques are used to evaluate the semantic relevance, clarity, and completeness of the answers. The proposed system supports multi-accent speech recognition to improve fairness and usability for diverse users. It also separates content evaluation and delivery analysis to provide transparent scoring and actionable feedback. The system is implemented using Python, Streamlit, Whisper, SentenceTransformers, and spaCy. Experimental testing shows that the system can effectively simulate interview practice, provide instant feedback, and help users improve their communication skills and confidence. The proposed solution is costeffective, accessible, and suitable for students, job seekers, and educational institutions.