A real-time interview assistant is a desktop app that listens to the interviewer's question during a live video or phone call, shows a caption of what was asked, and puts a structured answer on your own screen in about 2 seconds after the question ends. It differs from a chatbot because you never type the question in, and from a mock-interview tool because it runs during the actual interview instead of in a practice session beforehand.
That is the whole idea. Everything else on this page is detail: what the app captures, what it shows, who tends to use one, what it costs, and where it falls short. We make one, CoPilot Interview's real-time interview copilot, so the examples below describe that app specifically, but the definition applies to the category.
What "real-time" means in practice
The word covers three things that happen in sequence during a call on Zoom, Microsoft Teams or Google Meet:
- Capture. The app takes the interviewer's voice from your computer's audio, meaning whatever the meeting app plays, and mixes in your microphone for your own answers. Because the interviewer comes through the computer rather than the room, headphones work fine; a Mute mic toggle keeps only the interviewer's audio if you prefer.
- Captions. Live captions of the interviewer's words appear in the app's own window. This matters more than it sounds: you see the exact question, including the part you half-missed.
- A structured answer. About 2 seconds after the question ends, a written answer appears with the key points first and the detail underneath. Buttons let you ask for a shorter version, a longer version, bullet points, a redo, or type a question of your own.
The answer is on-screen text only. Nothing is played into your ear, and the app does not speak for you. You read the key points, decide what is true for you, and say it in your own words. If you want the mechanics in more depth, our guide to how real-time interview assistants work walks through the speech-to-text step by step.
How it differs from a chatbot and from a mock interview
People often lump all three into "AI for interviews." They solve different problems, and the table below is the fastest way to see the gap.
| Chatbot you type into | AI mock interview | Real-time interview assistant | |
|---|---|---|---|
| When it runs | Before the interview | Before the interview | During the live call |
| How the question gets in | You type it | The AI asks it aloud | Captured from the call's audio |
| What you get | A written answer | A score out of 10, feedback and a model answer | Captions and a structured answer in about 2 seconds |
| Best for | Drafting stories in advance | Rehearsing out loud and finding weak spots | A safety net on the day |
| Needs your microphone? | No | Yes, you answer out loud | Yes (mixed with the computer's audio; Mute mic available) |
A chatbot is a good prep tool, but in a live call you cannot type a two-sentence question while someone watches you. A mock interview tool is the practice counterpart: the AI mock interview builds questions from your pasted job description and résumé, reads each one aloud, and scores your spoken answer out of 10 with what worked, what to improve and a model answer. The real-time assistant is the only one of the three that is useful once the interviewer is actually talking. Most people end up using the second and third together: practice in the mock, then run the live assistant as a backup.
What it looks like on your screen
The app runs in a separate desktop window on your own screen, beside the meeting window. You do not have to click into it. Global hotkeys work while your cursor stays in the meeting app or a shared coding editor: Ctrl+Shift+L starts or pauses listening, Ctrl+Shift+A produces an answer, Ctrl+Shift+Q starts a new question, Ctrl+Shift+H shows or hides the window, and Ctrl+Shift+/ scrolls the answer. On a Mac, ⌘ replaces Ctrl.
You can also leave the hotkeys alone. Auto-answer can trigger after the interviewer pauses, and a new question is detected automatically after a 20-second pause in speech. The text size can be enlarged in Options, which helps if the window sits in a narrow strip near your camera.
Who uses one
The pattern we see most often falls into a few groups:
- Engineers in coding rounds. Coding mode returns complete code in your chosen language (50+ in the picker) with an example run, and a 🗣 Talk answer button handles a conceptual follow-up in words while Coding stays on. Read screen (Ctrl+Shift+S) captures a shared coding editor and answers what is shown.
- People interviewing in a second language. Captions and answers come in dozens of interview languages, and seeing the question written down removes the "did I hear that right?" problem.
- Candidates who freeze under pressure. Knowing a structure will appear if the mind goes blank lets some people relax enough that they rarely look at it.
- Hiring managers. A separate Interviewer Mode suggests next questions and produces a structured evaluation report, which is the same technology pointed the other way.
Behind the live window sits an AI interview assistant with more than the live mode: the mock interview, an interview report after each session with a scorecard and likely next-round questions, and a résumé builder.
What it costs
Prices as of October 2026, in USD. The Free plan is $0 forever, with no credit card and no signup: 15 minutes a day in sessions of up to 15 minutes, free AI models, and every mode including the mock interview and the résumé builder. Standard is $14.99 a month (regular $19.99) for sessions up to 60 minutes and up to 100 minutes a day with no waiting between sessions, plus Dual AI, where two models answer side by side. Pro is $29.99 a month with 70-minute sessions, unlimited sessions, and the ChatGPT and Gemini models. Pro+ is $49.99 a month and adds Claude. Annual plans and hour packs that never expire (1 hour for $5.99, 5 hours for $19.99, 15 hours for $34.99) are on the pricing page, along with the 14-day money-back terms for a first payment with under 30 minutes of live session time used. If you want to see how other tools in the category handle the same job, you can compare AI interview assistants side by side.
What a real-time assistant is not good for
It does not help in an in-person interview. There is no screen to glance at across a table, and the Android companion app (beta) listens through the phone's microphone for calls, not rooms. For those interviews the mock interview is the useful part of the product.
It also cannot know what you actually did. A suggested answer about "a time you handled conflict" is a structure, not your story. If you read a generic answer aloud, interviewers hear it. The app works best when your résumé and the job description are pasted in so answers reference your real projects, and when you treat every suggestion as a prompt rather than a script.
It needs an internet connection. Speech-to-text and the AI models run in the cloud, so it does not work offline. And it is not a recording tool: audio is streamed in short encrypted segments for transcription, and no audio, transcripts or answers are stored on CoPilot Interview's servers. Your session files land in your own Documents folder.
Is it allowed?
That depends on who is interviewing you and what they said in the invitation. Some employers now say openly that AI tools are fine; others ask you not to use them; many say nothing. Check the invitation and ask the recruiter if you are unsure.
Follow the rules of the employer or assessment you're in — many people use it only to practice, which works on the free plan.
See a real-time interview assistant on your own screen
The Free plan gives you 15 minutes a day, no credit card and no signup. Run it beside a practice call and watch the captions and answer appear.
Try it freeFAQ
Does a real-time interview assistant work on a phone call?
It works when the call's audio plays through your computer, for example through a softphone or a web dialer. If the call is on your phone, the Android companion app (beta) can listen through the phone's microphone when the call is on speaker, though it cannot capture call audio directly. There is no iPhone app yet.
Do I have to type the question in?
No. The app captures the interviewer's voice from your computer's audio and shows live captions, then produces the answer on its own after the interviewer pauses or when you press Ctrl+Shift+A. You can type a question of your own if you want to.
Does it read the answer into my ear?
No. Answers are shown as on-screen text in the app's own window, key points first. You read them and speak in your own words. The only thing the app reads aloud is the mock interviewer's questions during practice.
Can I try a real-time interview assistant for free?
Yes. CoPilot Interview's Free plan is $0 with no credit card and no signup, and gives you 15 minutes a day in sessions of up to 15 minutes with the free AI models. All modes are included, including the mock interview.
Which platforms does it run on?
It is a desktop app for Windows 10 and 11 and for macOS 12 and newer, downloaded as a .zip from copilotinterview.com or installed from the Microsoft Store. It runs beside Zoom, Microsoft Teams or Google Meet in its own window.
Does it store a recording of my interview?
No. Audio is streamed in short encrypted segments to a speech-to-text model and the question text goes to the AI model you picked, and no audio, transcripts or answers are stored on CoPilot Interview's servers. The transcript, answers and report are saved only in your own Documents folder.