A professional, state-of-the-art interactive mock interview simulator. This application leverages extremely fast open-weights LLMs via the Groq API and real-time audio manipulation to instantly deliver realistic, highly personalized mock interviews based on your specific resume!
- Interactive UI (Streamlit): Complete modern messaging-app layout mimicking a live chat with an expert HR Recruiter.
- Two Interview Modes: Toggle between Technical (evaluates deeply on architecture, languages, frameworks) or Behavioral/HR (evaluates on soft-skills and conflict resolution).
- Two-Way Audio Integration:
- Text-to-Speech (TTS): Fully integrated with Microsoft Edge's Neural TTS (
edge-tts). The AI physically speaks its questions out loud to you using a professional voice at a custom speed. - Speech-to-Text (STT): Built-in microphone support. Speak out your answers and the app instantly transcribes them using Groq's Whisper-Large-V3-Turbo model!
- Text-to-Speech (TTS): Fully integrated with Microsoft Edge's Neural TTS (
- Stateful Grading Loop: The AI dynamically processes your background, generates specific questions for you, listens to your answers in real time, grades them out of 10, and offers actionable resume-tailored feedback.
- Frontend Framework: Streamlit
- Core LLM Engine: LLaMA-3.3-70b-Versatile (via Groq SDK)
- Speech-to-Text: Whisper-Large-V3-Turbo (via Groq SDK)
- Text-to-Speech:
edge-tts(Microsoft Azure Neural TTS) - Document Processing:
pdfplumber,python-docx - Environment Management:
python-dotenv
- Clone the repository and ensure you have
uvorpipinstalled. - Install Dependencies (If using standard python):
pip install streamlit groq pdfplumber python-docx python-dotenv edge-tts gtts
- Configure Environment Variables:
Create a new file in the root folder specifically named
.envand paste your free Groq API key:GROQ_API_KEY=your-api-key-here
- Run the App!
Execute this command inside your terminal to launch the simulator in your browser:
streamlit run app.py
app.py: Handles Streamlit session states, constructs the interactive chat UI, and ties all backend functions together.prompt_generator.py: Interacts with Llama-3 using strict JSON schemas to generate custom technical/behavioral interview questions based solely on your resume background.interview_simulator.py: Acts as an AI grader that reviews your specific answer against the original question and determines what you missed.resume_parser.py: Robustly scrapes text arrays from your uploaded.pdfor.docxfile.audio_utils.py: Asynchronous helper class that handles high-speed Microsoft Neural text-to-speech rendering, and passes Streamlit audio captures into Groq's transcription algorithms.
This tool is strictly for professional development and educational practice. It does not predict real-life hiring outcomes.