InterviewSoul

Hear the question. See your answer.

InterviewSoul listens to your call on your own computer, writes down both sides, and puts a short answer on screen while you talk.

Your first 30 minutes are free. No card needed.

  • About 1 second to the first words of an answer
  • No audio leaves your computer
  • 25 languages understood

Whatever they ask, the answer is on screen before the silence gets awkward.

Three things happen during a call

  1. You press Start

    A small window sits on top of your call. It listens to your microphone and to what your computer is playing.

  2. It writes down both sides

    What they say and what you say appear as separate turns, with their questions marked as questions.

  3. An answer appears

    Press Answer, or switch on Auto, and a short answer you can say out loud shows up in about a second.

Built for glancing, not reading

You are mid-conversation. The answer has to be useful in the half second you can spare for it.

The first line is the one you say
Every answer opens with one or two sentences in your own voice, set larger than the rest. Supporting points follow if you need them.
It knows your background
Add your résumé, a CV and the job description once. Answers draw on your real experience and never invent any you don't have.
Your audio stays on your computer
Speech is turned into text on your own machine. Only the text of the conversation is sent to produce an answer, and we keep none of it.
Private while you share your screen
In private mode the window is left out of screen shares, recordings and screenshots. You can switch that off whenever you like.
Code you can use
For coding questions you get the approach, then the code with colours and a copy button, then the complexity. A screenshot button reads the problem off your screen.
The InterviewSoul window during a call: the conversation on the left and, on the right, an answer to a coding question with highlighted Python code. The Documents tab in Settings, with a résumé attached and a box for the job description. The Account tab in Settings, showing the Pro plan, the session time left this month and the renewal date.

You are mid-sentence, they are waiting, and you have half a second to look down. Everything here is built for that half second.

A full interview costs about three dollars

You pay for the hours you are actually in calls. A session is the time between Start and Stop, and nothing is counted outside one.

Free

$0

Try it on a real call first

  • 30 minutes of sessions, split however you like
  • Fast answers
  • Your résumé and job description
  • Screenshots and Auto answer
Download free

Basic

$12 a month

$2.40 for each hour of calls

  • 5 hours of sessions every month
  • Fast answers
  • Everything in Free
  • Cancel or switch in the app, any time
Start with Basic

Start on Free with no card. You choose or change a plan inside the app, under Settings, then Account, and cancelling keeps your plan until the end of the month you paid for.

Get InterviewSoul

Install it, sign in with your email, and press Start. The first time, it downloads what it needs to understand speech, about 490 MB.

Download for Mac Download for Windows Mac with Apple silicon, macOS 13 or later. Windows 10 or later, 64-bit.

Questions people ask first

Am I allowed to use this in an interview?

That depends on the rules of the interview, test or employer. Many do not allow outside help. Check before you use it; following those rules is your responsibility. InterviewSoul is also useful for practice interviews and ordinary meetings, where the question does not arise.

Does the other person know it is running?

Not from the app: it does not join the call or announce itself. It does write down what both of you say. Laws about recording or transcribing a conversation differ from place to place, and some require the other person's consent, so check what applies to you.

What leaves my computer?

Audio never does; speech is turned into text on your machine. When you ask for an answer, the recent text of the conversation, the text of your documents and, if you use it, a screenshot are sent to our server and on to OpenAI to write the answer. We keep a count of your session time, not what was said.

How fast is an answer?

The first words usually appear about a second after you ask. The fast models are the default for that reason; Pro lets you pick stronger ones, which think a little longer.

What happens when my time runs out?

Listening stops and the app tells you. Free time is 30 minutes once. Basic and Pro time comes back at the start of each month you pay for.

Which languages does it understand?

English and 24 other European languages, and it works out which one is being spoken. You can ask for answers in a particular language in Settings.