AI-Assisted Real-Time Interviewer
Budget: $30 – $250 USD
Project Description (for your freelance post)
I’m looking for a developer to build a real‑time AI interview assistant that listens to the interviewer’s voice during an online call (Zoom/Teams/Meet), transcribes it, and instantly generates AI‑based answers that the user can see in a web interface.
Core Requirements
Capture system audio (only the interviewer’s voice) on Windows.
Perform real‑time speech‑to‑text using an AI model (e.g., OpenAI Whisper API).
Stream the recognized text into a small web UI that shows:
Live captions of what the interviewer is saying.
Suggested AI answers to each question.
Use a chat model (e.g., OpenAI GPT‑4 family) to:
Understand the question.
Generate concise, professional answers.
Maintain conversation context across multiple questions.
Provide simple configuration for:
API keys.
Temperature / style of answers.
Target role (e.g., “Senior Full‑Stack Engineer”, “Data Scientist”).
Nice to Have
Optional text‑to‑speech to read the AI answer aloud.
Support for saving the full transcript + answers as a session log.
Basic prompt presets (technical interview, behavioral interview, product role, etc.).
Deliverables
Source code for backend + simple frontend (can be minimal, just functional).
Documentation on setup, running, and configuring the app.
Instructions for installing any audio capture dependencies on Windows.
I’m looking for a developer to build a real‑time AI interview assistant that listens to the interviewer’s voice during an online call (Zoom/Teams/Meet), transcribes it, and instantly generates AI‑based answers that the user can see in a web interface.
Core Requirements
Capture system audio (only the interviewer’s voice) on Windows.
Perform real‑time speech‑to‑text using an AI model (e.g., OpenAI Whisper API).
Stream the recognized text into a small web UI that shows:
Live captions of what the interviewer is saying.
Suggested AI answers to each question.
Use a chat model (e.g., OpenAI GPT‑4 family) to:
Understand the question.
Generate concise, professional answers.
Maintain conversation context across multiple questions.
Provide simple configuration for:
API keys.
Temperature / style of answers.
Target role (e.g., “Senior Full‑Stack Engineer”, “Data Scientist”).
Nice to Have
Optional text‑to‑speech to read the AI answer aloud.
Support for saving the full transcript + answers as a session log.
Basic prompt presets (technical interview, behavioral interview, product role, etc.).
Deliverables
Source code for backend + simple frontend (can be minimal, just functional).
Documentation on setup, running, and configuring the app.
Instructions for installing any audio capture dependencies on Windows.