A shared stage.
A personal translation.
- Speak from the stage
- Listeners join with a QR
- Read or listen to the translation
Live translation. In your browser.
Your words.
Everyone’s
language.
Speak your language. Your audience reads captions and hears a translation — with a link or QR code.
Explore for free after sign-up. Your available time is shown in your account.
Prepared example
Attuni.liveDemonstration
Attuni.liveDemonstrationOriginalEN / ES
Emma Wilson
Attuni translates live speech into text and reads the translation in the listener's language.
Diego García
Los subtítulos aparecen durante el discurso y se perfeccionan a medida que se reconocen.
TranslationES
Emma Wilson
Attuni traduce el habla en vivo a texto y lee la traducción en el idioma del oyente.
Diego García
Los subtítulos aparecen durante el discurso y se perfeccionan a medida que se reconocen.
Prepared audio · press to listen
For conversations in every setting
In the room and online
Different languages.
One conversation.
Live translation, on stage and on a video call.
A shared stage.
Alongside your everyday tools
Teams
Zoom
Google Meet
Slack
Discord
YouTube
Your meeting.
Across languages.
Open Attuni alongside your call.
- A language for everyone
- Real-time captions
- Translation in your headphones
01 / Speech recognition
Spoken words.
Living captions.
Captions appear as the conversation unfolds, refining until the phrase is complete.
- A separate source language for each speaker
- Names and colors keep the conversation clear
- Final turns are saved to history
Context, names and specialist termsRequires integration
ASR providers offer domain context, name hints and term lists. These are not connected to Attuni’s user interface yet. Automated diarization does not replace setting up separate speakers.
Source language: English
Emma Wilson
The next idea can come from anywhere. Let’s hear
ENGLISHLet’s build something together.
РУССКИЙДавайте создадим что-то вместе.
ESPAÑOLConstruyamos algo juntos.
TÜRKÇEBirlikte bir şeyler inşa edelim.
02 / Text translation
A language choice.
For every listener.
One turn, different translations. Listeners can switch language during the session and keep the original alongside.
126text translation target languages
Speech input and audio output have separate coverage from the text translation catalogue.
Language catalogue
Languages and regional variants come from the current product catalogue. The organizer chooses the targets for each room.
AcehneseAfrikaansAlbanianArabicAragoneseArmenianAssameseAymaraAzerbaijaniBashkirBasqueBelarusianBanglaBhojpuriBosnianBretonBulgarianBurmeseCantoneseCatalanCebuanoSimplified ChineseTraditional ChineseChineseCroatianCzechDanishDariDutchEnglishAmerican EnglishBritish EnglishEsperantoEstonianFinnishFrenchCanadian FrenchFrench (France)GalicianGeorgianGermanGerman (Germany)Swiss High GermanGreekGuaraniGujaratiHaitian CreoleHausaHebrewHindiHungarianIcelandicIgboIndonesianIrishItalianJapaneseJavanesePampangaKazakhKonkaniKoreanKurdishCentral KurdishKyrgyzLatinLatvianLingalaLithuanianLombardLuxembourgishMacedonianMaithiliMalagasyMalayMalayalamMalteseMāoriMarathiMongolianNepaliNorwegian BokmålOccitanOromoPangasinanPashtoPersianPolishBrazilian PortugueseEuropean PortuguesePortuguesePunjabiQuechuaRomanianRussianSanskritSerbianSouthern SothoSicilianSlovakSlovenianSpanishEuropean SpanishLatin American SpanishSundaneseSwahiliSwedishFilipinoTajikTamilTatarTeluguThaiTsongaTswanaTurkishTurkmenUkrainianUrduUzbekVietnameseWelshWolofXhosaYiddishZulu
Context and terminology
Available: translation receives previous final turns from the same speaker as context. Custom glossaries, formality and your own instructions require integration.
Available Previous-turn context · Requires integration Custom glossary
03 / Spoken translation
Read the words.
Hear the meaning.
Turn on translated audio, choose a comfortable speed and follow the highlighted phrase. Sound when you want it.
- Play at 1×, 1.25×, 1.5× or 2×
- See the current turn highlighted as it plays
- Voices from the room’s available configuration
Voices, delivery and pronunciationRequires integration
Voices are configured on the server; listeners see a selector only when several compatible voices are available. Cloning, emotions, pronunciation dictionaries and creating your own voice are not connected to Console.
Prepared example
Attuni translates live speech into text and reads the translation in the listener's language.
Prepared audio · press to listen
Your room. Your controls.
Create a room.
On your own terms.
Speakers, languages and a public link, together in one interface.
International forum
forum.attuni.live/annual-meetingDemonstrationSpeakers
EWEmma WilsonEnglish
DGDiego GarcíaSpanish
Translation languages
EnglishRussianTürkçeFrench
Console interface reproduction · account limits apply
One link for everyone
https://room.attuni.live/This QR opens the public demo room.
Speaker microphoneEmma Wilson · EN
Audio input available to the browser
Speakers talk. Listeners follow.
In an active room, each speaker opens an invitation, selects an audio input and manually turns on their microphone. The organizer controls start, pause and finish.
Try for freeView actual product screens
Scan to open the demo
Attuni.live DemonstrationOriginal
Emma Wilson
Welcome to our conversation.
Translation · EN
Emma Wilson
Welcome to our conversation.
The listener’s experience
Open a link.
Join the conversation.
No app to install. Pick a language and make the screen work for you.
- Original and translation together, or one feed
- Earlier turns and a return to the live feed
- Light and dark themes
- A floating window in supported browsers
Audience settings example. A floating window requires Document Picture-in-Picture browser support.
Explore the demoWhere understanding matters
In the room. Online.
Across your team.
One browser-based workflow for different kinds of conversations.
Conferences
Put the QR on screen. Each listener picks a language on their own phone.
Webinars
Share the translation link with your invitation and offer both text and audio.
Business meetings
Set each speaker’s language and follow the original alongside the translation.
Training
Help participants read explanations and revisit earlier turns.
Live streams
Route an audio input to the browser and share a translation link with viewers.
Hybrid events
One room connects people in the venue with those joining remotely.
ZoomMeetTeamsYouTube
Open Attuni alongside your usual service. Speakers select a browser-accessible audio input; system audio needs separate routing. Native integrations, meeting bots and automatic call capture are not claimed.
After the conversation
The meeting ends.
The words stay.
Keep originals and translations. Revisit room sessions and see how your audience connected.
Saved conversationPrepared example
Emma Wilson · ENAttuni translates live speech into text and reads the translation in the listener's language.
Translation · ENAttuni translates live speech into text and reads the translation in the listener's language.
JSON and TXT after the session ends and data is saved. Export everything, one language or one speaker. SRT/VTT are not available yet.
Audience languagesExample data
128Visitors
86Peak online
English64
Русский38
Türkçe26
Illustrative values. Console also covers devices, browsers and audio usage.
Attuni for your brand
Your event.
A familiar experience.
Today: your event name, address and speaker colors. Tailored branding is shown as a concept to discuss.
Standard roomAvailable
Attuni.liveInternational forum
forum.attuni.live/annual-meeting
Custom branding conceptConcept
Your company forum
Ideas that bring us together.
Example: translate.company.example
Available
Event title, speaker names and colors; your own subdomain and path within attuni.live.
Requires integration
Your logo and brand palette are a concept. An arbitrary domain requires separate implementation; Console has no white-label editor.
Around the conversation
Everything around
a live session.
Session controls
Start, pause and resume your room. Follow its state and review previous sessions.
A familiar interface
Console and Speaker have 10 interface languages, including Arabic with right-to-left layout.
API and deployment
The project has an API and OpenAPI. A self-hosted installation or a custom workflow requires integration.
Speech, translation and voice technology
SonioxASR→DeepLMT→InworldTTSAvailable
Attuni assigns the current pipeline. Console does not offer arbitrary provider selection.
Terminology, voices and other extensionsRequires integration
| Capability | Status |
|---|---|
| Custom ASR context and glossaries | Requires integration |
| Custom voices, emotions and cloning | Requires integration |
| Files, AI summaries and meeting assistant | Requires integration |
Retained adapters for Deepgram, Google Translation, Azure Translator, ElevenLabs, Cartesia, Gemini TTS and OpenAI do not make their features available to users.
These are possible development directions, not active services or delivery commitments. A model’s capabilities and provider terms need checking before integration.
Let every listener
hear your idea.
Get to know Attuni in the demo, or create a room of your own.
Questions and answers
A few things
to know.
How to get started, and what to expect from the current version.
Contact via TelegramDo I need to install anything?
Organizers, speakers and listeners use a browser. Microphone access needs browser permission. Continuous capture or playback with a locked mobile screen is not guaranteed.
How do I create a room and invite people?
Open the app, name your room, add speakers and choose translation languages. Send speakers their invitations, and listeners the public link or QR. Available time and limits are shown in your account.
Which languages are supported?
The current catalogue has 126 text translation targets. Speech recognition and audio output have separate lists and restrictions. Each speaker has an explicitly selected source language; the organizer chooses the room’s targets.
Can I read and listen at the same time?
Yes. Audience shows the original and translation and lets you turn on spoken translation. Playback speeds are 1×, 1.25×, 1.5× and 2×. Available voices depend on the room configuration. This is synthesized translation, not a copy of the speaker’s voice.
How does it work with Zoom, Meet, Teams or YouTube?
Attuni opens alongside the service in your browser. The speaker selects an audio input; capturing system audio may need separate routing. Native integrations, meeting bots and a virtual microphone are not part of this workflow.
Can several speakers take part?
Yes. Each has an invitation, name, color and explicitly selected source language. The organizer manages speakers; the number available depends on the account.
Where is history, and what can I export?
Listeners can revisit earlier turns. Owners see room sessions and export saved originals and translations as JSON or TXT after a session ends. SRT/VTT are not available yet.
How can I try Attuni for free?
The public demo needs no registration and uses prepared turns and audio. To use your own room, register in the app: your current trial balance and limits are displayed in the account. The balance belongs to the account, not each room.
Can I use my logo and domain?
Event titles, subdomains and paths within attuni.live, speaker names and colors are available. Your own logo and arbitrary palette are shown as a concept. A custom domain requires infrastructure work; there is no self-service Console flow.
Are glossaries and domain context available?
Translation uses previous final turns from the same speaker. Custom ASR hints, glossaries, style and pronunciation dictionaries require integration and are not available Console settings.
Which browsers support the floating window?
It is available when the browser supports Document Picture-in-Picture. Support varies; check for the corresponding Audience or Speaker control. The regular translation screen works without it.
What is known about data and privacy?
Final originals and translations are saved for history and export. Raw analytics details have a 30-day retention after completion; anonymized daily aggregates remain. This is not a promise of complete deletion, certification or a particular processing jurisdiction. Discuss your requirements before sharing sensitive data.