--- title: AI Interview Coach emoji: πŸŽ™οΈ colorFrom: blue colorTo: indigo sdk: streamlit app_file: app.py pinned: false license: mit short_description: "Real-time voice interview simulation powered by Llama-3" --- # πŸŽ™οΈ AI Interview Coach: Real-Time Voice Simulator This project is an **Autonomous Interview Agent** designed to conduct realistic, voice-based technical interviews. It processes a candidate's Resume (PDF), generates context-aware technical questions, and interacts via voice with ultra-low latency. **Built by:** Merve Γ‡alışkan **Purpose:** To demonstrate an end-to-end Generative AI pipeline featuring RAG, Audio Processing, and LLM orchestration. --- ## πŸš€ Key Features * **πŸ“„ RAG-Based Context:** Analyzes uploaded PDF resumes to generate tailored, domain-specific interview questions. * **⚑ Ultra-Fast Inference:** Powered by **Groq API** running **Llama-3-8b**. * **πŸ—£οΈ Speech-to-Text (STT):** Utilizes **Whisper-large-v3** (via Groq) for highly accurate transcription. * **πŸ”Š Text-to-Speech (TTS):** Converts the AI's feedback and follow-up questions into audio. * **🧠 Chain-of-Thought Evaluation:** The AI listens, evaluates logic, and provides feedback. --- ## πŸ› οΈ Tech Stack * **LLM Engine:** [Groq Cloud](https://groq.com/) (Llama-3-8b-8192) * **Speech Processing:** Whisper-large-v3 (STT) & gTTS (TTS). * **Framework:** [Streamlit](https://streamlit.io/) * **Orchestration:** LangChain * **Vector/Data Handling:** PyPDFLoader --- ## βš™οΈ How It Works 1. **Upload Resume:** The user uploads a PDF resume via the sidebar. 2. **Context Extraction:** The app extracts text from the PDF. 3. **Voice Interaction:** User speaks, AI transcribes, thinks, and responds with voice. --- ## πŸ“¦ Local Installation To run this project locally: 1. **Clone the repo:** ```bash git clone ``` 2. **Install dependencies:** ```bash pip install -r requirements.txt ``` 3. **Set up API Keys:** Create a `.streamlit/secrets.toml` file: ```toml GROQ_API_KEY = "your-groq-api-key-here" ``` 4. **Run the app:** ```bash streamlit run app.py ```