macOS · thirteen languages · nothing leaves your Mac

Talkback you can read.

Monitors, stage and production, each on their own line, each in their own colour. Read what you missed instead of asking them to say it again.

HUUTO — TavastiaRECORDING
MonitorsCan we get more vocal in wedge three21:04
StageGuitar DI is dead, patching the spare21:04
ProductionDoors in five, hold the soundcheck21:05
MonitorsdraftThat is better, thanks21:05
FOHCopy that — holding.21:05
An illustration of the app's chat view, built from the same theme values and interface strings as the app itself. Not a screenshot.

The problem

Talkback is the channel you can't miss and can't hear.

Monitor world needs a word. Production is asking something mid-song. Someone at the patch wants to know which line went dead. It all arrives in the same ear, under a band at full tilt, in half-sentences and first names — and you catch maybe half of it.

HUUTO puts the same thing on screen. Not instead of the talkback bus, but alongside it, as a copy you can read.


How it works

Three steps, then you forget it is there.

  1. Plug in what you already have

    Dante, MADI, AVB, SoundGrid, a USB mixer — or nothing at all. The Mac’s own microphone works for trying it out.

  2. Name the channels

    Point each input at a person. Monitors, stage, production. Give them a colour and group them however your show is organised.

  3. Read the talkback

    Speech turns into chat as it happens. Search it, star the lines that matter, and take the whole show home as a PDF.

Everything happens on your Mac. The recogniser is the one macOS already ships, so there is no model to download and nothing to send anywhere.


What it does

Built for the desk you are already standing at.

Works with your rig

Every audio interface with inputs shows up in the list. No brand is left out, and you need no extra hardware to try it.

Text appears while they talk

A first draft appears while the talker is still talking, then sharpens up as the sentence lands. Every channel gets one, not just the loudest.

One line per person

Every channel has a name and a colour. Group them into monitors, stage and production — a closed group still shows you who is talking.

Share it with the crew

Anyone on the same network can follow along on another Mac, and type back to FOH from there.

Cue words that reach the desk

Get a banner the moment your word is spoken, and send an OSC message to lighting or media from the same hit.

Every show, saved

Sessions are kept automatically. Search by word, filter by person, star what matters, export to text or PDF.


Languages

Thirteen languages.

Choose the language in settings before the show. HUUTO keeps it for the whole session, so nothing drifts halfway through the set.

  • English
  • Finnish
  • Swedish
  • Norwegian
  • Danish
  • German
  • Dutch
  • French
  • Spanish
  • Italian
  • Portuguese
  • Polish
  • Estonian

These are the languages HUUTO is set up for. Talkback is short, overlapping and half-shouted, which is a harder job than clean dictation — so the list is the ones that hold up.


Stays on your Mac

Nothing is sent anywhere.

  • No cloud, no account, no subscription. The speech is recognised on your own machine.
  • Venue wifi can be down and it keeps working. The only time HUUTO needs the network is if macOS does not yet have the dictation files for your language.
  • No recordings are kept. Only the text you see is saved.
  • Recognition uses your Mac’s own speech engine, never Apple’s servers. macOS manages the languages; if yours is missing you can add it in System Settings.

Download

Coming soon.

It will come straight from this page and open like any other app. No App Store, and no account to create.

What you need

  • A Mac running macOS 26 or newer
  • Apple Silicon
  • Any audio interface with inputs, or just the built-in microphone to try it

Worth knowing

Where the limits are.

  • It is not an intercom

    HUUTO does not replace talkback, intercom or any other channel you rely on when it matters. Keep the one you have — this reads alongside it.

  • A good copy, not a record

    Names, numbers and call signs come out wrong sometimes. Close enough to follow a show by, not close enough to quote from.

  • Sixteen channels

    Sixteen is the ceiling. How many run comfortably at once depends on your Mac and how much talking there is, so try your real setup before the show.

  • Names come from channels

    Each channel carries the name you gave it, so every line is attributed the moment it appears. Two people sharing one microphone share one name.

  • Sharing is unencrypted

    Anyone who can reach the network can read the transcript and write to it. Keep it on the production network, not on venue or hotel wifi.