metadata
title: Smart Interview Simulator
emoji: 🎤
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 6.5.1
python_version: '3.10'
app_file: app.py
pinned: false
🎤 Smart Interview Simulator (AI Voice Bot)
An AI-powered voice-based mock interview platform built using Speech Recognition and Large Language Models.
This project allows users to simulate a technical interview by answering questions using voice input and receiving structured AI-generated evaluation and feedback.
🚀 Features
- 🎙 Voice-based answer submission
- 🧠 Speech-to-Text using Whisper
- 🤖 AI answer evaluation using Mistral-7B-Instruct via Hugging Face Inference API
- 📊 Structured scoring:
- Technical Accuracy
- Clarity
- Depth
- Overall Score
- 🎯 Difficulty levels (Easy / Medium / Hard)
🏗️ System Architecture
User Voice Input
↓
Speech Recognition (Whisper)
↓
Answer Transcription
↓
Prompt Engineering
↓
LLM Evaluation
↓
Structured Feedback & Score
🛠️ Tech Stack
- Python
- Gradio
- Transformers
- Hugging Face Inference API
- Whisper (Speech Recognition Model)
- Mistral-7B-Instruct (Large Language Model)
📊 Evaluation Criteria
The AI evaluates answers based on:
- Technical Accuracy (0–10)
- Clarity (0–10)
- Depth (0–10)
- Overall Score (0–10)
- Improvement Suggestions
🔮 Future Improvements
- 📈 Performance analytics dashboard
- 💾 Interview history tracking
- 📄 PDF performance reports
- 🎙 Emotion detection from voice
- 🔄 Follow-up question generation
- 🧑💻 Coding interview simulation
🏆 Resume Description
Developed a Voice-Based AI Interview Simulator using Whisper for speech recognition and Mistral-7B via Hugging Face Inference API for intelligent technical answer evaluation and structured scoring.
📜 License
This project is intended for educational and research purposes.