Voice Mode: Listen to Answers Without Reading the Screen

Reading an answer on screen while maintaining eye contact is a skill most people do not practice. In a high-pressure interview, your eyes drift, your reading speed drops, and the interviewer notices. MentionBell Voice mode removes that problem entirely — it reads your answer aloud through text-to-speech so you hear it instead of reading it.

How Voice Mode Works

When Voice mode is enabled, every answer MentionBell generates is spoken through your audio output after the text is ready. The flow is:

  1. Interviewer asks a question

  2. MentionBell detects the question on partial speech and generates a CV-grounded answer

  3. The answer appears on screen (Standard or Stealth) and is read aloud

  4. You hear the key points in your ear while keeping your eyes on the camera

Voice mode is a delivery layer — it works with both Standard (on-screen overlay) and Stealth (phone display). You can also use Voice alone with Stealth hidden, so neither laptop nor phone shows text.

When Voice Mode Helps Most

Panel interviews

Three people watching your face will notice if you look down at a screen. Voice mode lets you hear the answer structure while your gaze stays forward.

Behavioral rounds with follow-ups

Behavioral answers are long. Reading a full STAR response from a screen takes 10–15 seconds of visible eye movement. Hearing it takes the same time but looks natural — like you are collecting your thoughts.

Phone screens

There is no camera to maintain eye contact with, but there is also no screen to read from if you are on a phone call. Voice mode delivers the answer through your earpiece while you speak.

Combined with Stealth

For maximum discretion: Stealth mode puts answers on your phone, Voice mode reads them in your ear. You glance at nothing. You read nothing. You hear the answer and speak it in your own words.

Setting Up Voice Mode

  1. Open MentionBell and go to session settings

  2. Enable Voice mode

  3. Connect a single earbud or earpiece to your audio output

  4. Leave the other ear open so you can hear the interviewer naturally

  5. Start your session and test with a practice question

Tip: Test the volume before the real interview. You want to hear MentionBell clearly without the interviewer hearing the TTS through your open microphone. A single wireless earbud with low volume works well.

Voice Mode vs Reading: A Practical Comparison

  • Eye contact — Reading requires practiced downward glances; Voice maintains it naturally

  • Answer length tolerance — Reading suits short to medium answers; Voice handles long behavioral answers well

  • Setup complexity — Reading is minimal; Voice requires an earpiece

  • Coding rounds — Reading is better (you need to see code); Voice is less ideal

  • Discretion — Reading is high with Stealth; Voice is highest with Stealth + Voice

For coding rounds, stick with Standard mode and the screenshot shortcut — you need to see approach, complexity, and code on screen. For everything else, Voice mode is worth trying.

Custom Instructions Still Apply

Voice mode does not change how answers are generated. Your custom instructions — STAR format, tone, question classification — all apply before the text-to-speech step. The spoken output follows the same structure as the written answer.

If your instructions say "keep answers under 60 seconds when spoken," MentionBell respects that in both text and audio.

Common Questions

Can the interviewer hear the TTS?
Not if you use a single earbud at low volume and your mic is not picking up your earpiece. Test this in a mock call.

Does Voice mode slow down answers?
The text generates at the same speed (~2 seconds to first words). TTS playback starts as soon as the text is ready — typically a fraction of a second after.

Can I turn Voice on mid-session?
Yes. Toggle it in settings without restarting the session.

Recap

  • Voice mode reads CV-grounded answers aloud via text-to-speech

  • Best for behavioral rounds, panel interviews, and phone screens

  • Combine with Stealth for maximum discretion — hear answers, see nothing

  • Use Standard + screenshot for coding rounds where you need visual output

  • Custom instructions apply to spoken answers the same as written ones

Try MentionBell — unlimited sessions on every plan. Download · See pricing

Related reading: