AI models and routes

The right route per language. Not one model for everything.

Pick a route per language. Tõlk keeps it running and switches engines if one fails. Availability differs by product and language.

Direct speech-to-speech

A realtime model listens and speaks the translation directly, keeping the speaker’s pace.

Speech in
Realtime translation model
Translated speech out

Best for

Keynotes and talks
Fast, natural delivery
Many languages at once

Trade-offs

Names and jargon follow the model’s own understanding
Voice is the model’s own

Engines and voices

OpenAI realtime translation · Default engine
Gemini Live Translate · Widest language list
Qwen LiveTranslate · Opt-in; processed outside the EEA

Conference Spotlight Polylogue Bridge

Dictionary-assisted

Speech is recognized, translated with your event Dictionary, then read by a chosen voice.

Speech in
Recognition
Translation with your terms
Voice

Best for

Names, brands and terminology
Consistent voices per speaker
Specialist seminars

Trade-offs

Slightly more delay than direct translation
Works sentence by sentence

Engines and voices

ElevenLabs, OpenAI or Gemini recognition
OpenAI translation with your Dictionary
ElevenLabs or OpenAI voices

Conference Spotlight Bridge

Prepared and reviewed

For recordings: translate and voice ahead of time, review, then broadcast or export.

Recording
Transcribe + translate
Voice per speaker
Human review

Best for

Recorded talks and courses
Scheduled premieres
Content you will reuse

Trade-offs

Not live: needs preparation time

Engines and voices

Premium studio voices · ElevenLabs
Standard voices · OpenAI or Gemini

Studio

Providers

Four providers. One setup.

We bring the best mix of models so every language gets high-quality translation. You never open a provider account or connect a key. Let Tõlk pick, or take every decision yourself.

Let Tõlk pick

Not sure which model fits? Every product starts on a tuned setup per language, with a backup engine ready.

Best engine per language, chosen for you
Automatic failover and switch-back
No provider accounts or API keys

Take full control

Care about the details? Set the route, engine, recognition model and voice yourself, language by language.

Engine and route per language
Recognition model per event
Voice and speed per language or speaker
Your Dictionary for names and terms

Another model in mind? We’ll bring it in.

Talk to us about your models

Another providerA vendor or model you already use or trust.
Open-source modelsOpen-weights speech, translation or voice models.
Self-hostedModels on your own servers or private cloud.

How a route is chosen

What matters for your content.

Language coverage

Engines support different language lists. Tõlk shows which engines can serve each language.

Delay

Direct routes start speaking sooner; dictionary routes trade a little time for precision.

Terminology

When names and terms must be exact, a Dictionary-assisted route uses your glossary.

Voice

The model’s own voice, or a chosen voice per speaker.

Region and privacy

Some engines process outside the EEA; they are opt-in and labelled.

Resilience

If an engine fails, languages switch to a backup automatically and switch back on recovery.

Voices

Studio-voice and dictionary routes read translations in ElevenLabs or OpenAI voices, set per language, and in some products per speaker.

Technical documentation Test routes on your content

The Live AI Translation Hub

Ready for your next moment?

Tell us what to make multilingual. We pilot it on your content, then you decide.

Request a pilot Talk to the team · hello@tolk.fm