Back to Catalog

Build Your Interview Coach with Llama, OpenAI Whisper & TTS

IntermediateGuided Project

Build an LLM-powered interview coach that analyzes resumes, reads job descriptions, and conducts personalized mock interviews with OpenAI Whisper speech tool. Extract and summarize resume content automatically, interpret job requirements, and generate tailored interview questions that adapt to the user’s responses in real time. Convert interviewer prompts into natural-sounding audio, record the user’s spoken answers, and transcribe them with Whisper for deeper evaluation. By the end, you’ll build a web app simulating a realistic interview experience, helping you ace interviews!

Language

  • English

Topic

  • Artificial Intelligence

Skills You Will Learn

  • LLM, AI Agents, OpenAI Whisper, Gradio, Generative AI, Python

Offered By

  • IBMSkillsNetwork

Estimated Effort

  • 45 minutes

Platform

  • SkillsNetwork

Last Update

  • December 4, 2025
About this Guided Project
Have you ever wished you could sit down for a realistic job interview, which is tailored exactly to your resume and the role you want, without scheduling with a recruiter, asking a friend for help, or feeling the pressure of a real interview?

Interview preparation is one of the most important steps in landing a job, but it’s often stressful and uncertain. What questions will they ask? How do your past experiences align with the role? How can you practice speaking confidently and clearly? Traditional prep tools rarely adapt to your personal background, and most mock interviews feel scripted or generic.

What if an AI agent could read your resume, understand the job you’re applying for, and conduct a personalized, dynamic interview? In this project, you’ll build an AI-powered interview coach that does exactly that. You’ll integrate resume parsing, job-description analysis, natural language generation, speech synthesis, and speech-to-text transcription into one seamless pipeline. Your system will read PDFs, summarize key strengths, generate tailored interview questions, convert them into natural interviewer audio, record the user’s voice responses, transcribe them, and continuously evaluate performance. Everything culminates in a polished, interactive Gradio web interface that lets anyone rehearse for a job interview anytime, anywhere.

What You’ll Learn

By the end of this project, you will be able to:
  • Extract and understand resume content automatically: Use PyPDF2 to parse resume PDFs and generate structured summaries to inform the interview.
  • Analyze job descriptions with LLM reasoning: Build a job description expert agent that interprets role expectations, required skills, and hiring priorities.
  • Generate intelligent, contextual interview questions: Use a powerful LLM to ask adaptive questions based on both the resume and prior answers.
  • Convert interviewer prompts into speech: Implement text-to-speech with gTTS to create natural-sounding, interactive audio-based interviews.
  • Capture and transcribe user responses: Use the Faster Whisper model for high-quality speech-to-text transcription of microphone input.
  • Deliver real-time evaluation: Build a custom evaluation agent that summarizes performance, strengths, and areas for improvement.
  • Develop a full Gradio app: Create a smooth, shareable interface that ties all components together—upload, listen, respond, and receive feedback.

Who Should Enroll

This project is perfect for:
  • AI enthusiasts and developers interested in combining language models, audio processing, and user-facing interfaces.
  • Job seekers and career coaches who want to harness AI to create personalized interview training tools.
  • Students and researchers looking to build multi-step AI systems that integrate text, speech, and interactive workflows.
  • Educators who want a hands-on project demonstrating real-world applications of LLMs and voice interfaces.

Why Enroll

This project brings the worlds of career development, conversational AI, and multimodal interaction together. Instead of guessing what interviewers might ask—or searching through lists of generic practice questions—you’ll learn to build a tool that understands your background and adapts to your needs. With your AI interview coach, every question, follow-up, and evaluation will be custom-made for the role you’re targeting.
By the end, you’ll have:
  • A fully functional AI interview simulation tool.
  • Practical experience integrating text, audio, and LLM workflows.
  • A deeper understanding of how generative AI can transform professional skill-building.
  • A portfolio-ready project demonstrating real-world relevance and engineering depth.

What You’ll Need

To get the most out of this project, you should have:
  • Basic Python programming skills.
  • Some familiarity with AI concepts (LLMs, speech processing, etc.)—helpful but optional.
  • Curiosity about how generative AI can enhance professional training and personal development.
All required libraries are installed in the environment, and the web app works best on the latest versions of Chrome, Firefox, Edge, or Safari.

Instructors

Jianping Ye

Data Scientist Intern at IBM

I'm Jianping Ye, currently a Data Scientist Intern at IBM and a PhD candidate at the University of Maryland. I specialize in designing AI solutions that bridge the gap between research and real-world application. With hands-on experience in developing and deploying machine learning models, I also enjoy mentoring and teaching others to unlock the full potential of AI in their work.

Read more

Contributors

Joshua Zhou

Data Scientist

I like building fun and practical things.

Read more

Zikai Dou

Data Scientist at IBM

Ph.D. Candidate in Computer Science at McMaster University, specializing in Federated Learning (FL), Graph Neural Networks (GNNs), and Computer Vision (CV). I develop privacy-preserving, distributed AI systems that tackle real-world challenges in healthcare, finance, and enterprise applications. Passionate about bridging academic research with industry impact to advance scalable and trustworthy AI.

Read more