Voices from several providers
Azure, ElevenLabs, Gemini, and more — behind a single interface with no provider-specific code.
The voice
AetherTongue
Natural, multilingual text-to-speech for the arena. One interface in front of several voice providers — every written line becomes a clear announcement.
Overview
A voice service that speaks the way the hall needs
Arena audio lives and dies by the voice that reads it. AetherTongue gives the platform one interface to the best available text-to-speech voices, so a written line becomes a clear announcement — without your team managing several TTS accounts and keys.
The problem
Robots should not read the lineup
Text-to-speech is easy to bolt on and hard to do well. Most setups sound robotic, or force you to juggle accounts:
- A robotic voice drains the atmosphere from the hall exactly when it matters.
- Several TTS accounts and keys create admin work nobody signed up for.
- One language rarely covers the mix of visitors an arena actually has.
Key features
What AetherTongue gives the platform
Multilingual by default
Announce a line in the languages the moment needs, without a separate configuration per voice.
Clarity before personality
We listen and curate voices that are natural but still cut through the noise of a live hall.
One line, any voice
Switch the voice per language or per venue — the written line never has to change.
Use cases
Where AetherTongue is heard
Match-day lineups
Player names and starting fives read clearly over the speakers, in the languages the club needs.
Pregame scripts
Welcome messages and event notes generated once and spoken consistently at every home game.
Countdowns and calls
Short, timed announcements that keep the hall informed without a second microphone.
Who it is for
Built for anyone who has to be understood
Sound operators
Operators who want announcements without typing them out live every time.
Clubs with mixed audiences
Venues where visitors speak several languages and expect to understand what is said.
Accessibility-minded venues
Halls that want spoken information as a complement to screens and signage.
FAQ
AetherTongue questions
Swedish and English today, with more added as our voice providers support them.
If your voice is available on a provider we support — yes, bring your own.
Announcements are generated when you need them. We do not keep recordings you did not ask for.
Provider voices vary. We curate the voices we expose — listen to a sample and judge for yourself.
AetherTongue works best with
Let every announcement sound like it belongs
Tell us which languages your hall needs, and we will show you what AetherTongue can do today.