Try it now
A working sandbox. No sign-up, no project. Sample data only.
- Multiple translation engines available. Pick the one that gives the best tone for your language pair.
- 20+ target languages including Spanish, French, German, Japanese, Mandarin, Arabic, Hindi.
- Translate during the live session or as a background pass on an existing transcript.
- Translator-tuned prompts preserve speaker voice rather than over-polishing.
Translation will appear here.
How it works
Three steps from raw material to result.
From the language picker on a live session, a file, or a live-share viewer. Each viewer can pick independently.
We knew by the second week.
Lo supimos a la segunda semana.
The numbers stopped making sense.
Los números dejaron de tener sentido.
Each line is queued to the translation engine as it's transcribed. The translated column updates within a second of the source.
Change the target on an existing transcript and a background batch re-translates without re-running transcription.
Frequently asked questions
Which languages are supported?
Spanish, French, German, Portuguese, Italian, Dutch, Polish, Russian, Ukrainian, Turkish, Arabic, Hebrew, Persian, Hindi, Bengali, Tamil, Mandarin, Cantonese, Japanese, Korean, Vietnamese, Thai, Tagalog, and Indonesian, plus several more. The full list is visible in the language picker.
Which translation engine runs by default?
Google Cloud Translation when a Google key is available; otherwise OpenAI. You can pick DeepL for European-language interviews where its tone tends to be tighter.
Does translation cost extra?
Live translation is metered per minute in addition to live transcription. File translation is metered per minute of source audio. Rates are on the pricing page.
Can I translate an old interview?
Yes. Open any transcribed file, pick a target language, and translation runs as a background batch. The result lands as a second column next to the source.
Will the translation preserve names and quotes?
Proper nouns, place names, and direct quotes are preserved with a translator-style approach that prioritises the speaker's voice over editorial polish.
Related capabilities
Further reading
Background guides and comparisons.
How Google Translate, DeepL, and OpenAI translation handle tone, names, and idioms differently, and why the engine that works for corporate content is often wrong for documentary speech.
How to handle proper nouns, speaker voice, and over-polish when sending documentary transcripts through machine translation, and when to flag a segment for a human reviewer.