Online Vocal Interviewer: based on AI for Candidate Response Analysis and Evaluation

This paper presents an implementation of AI-powered Voice Based interviewer. It includes a dynamic generation of interview questions from the given job profile description, conversational interaction in real time, and a scoring system via Sentiment Analysis based on structural coherence and content relevance. A web dashboard aggregates the scoring and statistics with a summary of the whole interview. The core tech stack used for this project is Next.js, Supabase, and Retall AI. For sentiment analysis OpenAI API isorchestrated through LongChain framework.