Fifteen languages, out of one audio
Somebody speaks once and the room hears it in fifteen languages. Not fifteen passes over the recording — one listen, and every language comes out of it.
Fifteen languages from the same audio, at once
The expensive part is understanding what was said, and that happens once: every language comes out of that single listen. That is why you can have all fifteen open without setting up fifteen rooms or fifteen booths. On the invoice each language costs the same —2 credits per spoken minute— and they all count, the one being spoken included: a meeting in Spanish heard in English, French and German is four.
| Languages | What a spoken minute costs |
|---|---|
| 1 language | 2 credits per spoken minute |
| 3 languages | 6 credits per spoken minute |
| 5 languages | 10 credits per spoken minute |
| 15 languages | 30 credits per spoken minute |
Charged on speech, not on how long the file is. An hour of recording with twenty minutes of talking pays for twenty minutes, whether it goes out in one language or fifteen.
Male or female, per language
Each target language is set to a male or a female voice, and it is a setting on the room rather than a decision we make for you. Change it and the next connection uses it.
Speed, per language
Some languages take longer to say the same thing and their audio piles up behind the speaker. Raising the speed of that one language brings it back in line, and it is set on that language alone.
Any of them can be the one spoken
The fifteen work as the language spoken in the room and as a language it is heard in. A Ukrainian speaker heard in Spanish is the same job as the other way round.
The site is in thirteen — that is a different list
You are reading this in one of thirteen languages, which is how many the website is written in. What the engine translates is these fifteen. We keep the two apart on purpose: promising a language we cannot speak is the worst thing a page like this can do.
The one you need is not here
Tell us which, and what you would use it for. Adding a language is not a switch — it needs a voice that sounds like a person and enough testing to trust it — but knowing somebody is waiting for one is what puts it in the queue.
The words only used in your house
An organisation's own names — people, streets, references, job titles — are exactly what a general model has never heard, and where it fails: it swaps them for an ordinary word that sounds similar. Write them once and they stop failing. You write them in the room and they hold for all its meetings.
- It is a list, not training: you write it and it works from the first sentence.
- It collects nothing. Not your audio, not your text, not one second of anything.
- And it is included. Not an extra plan, not a line on the bill.
The jargon of your trade
The engine hears everyday language well. It fails on the vocabulary used only inside a profession — «paroxysmal nocturnal dyspnoea», «plea of lis pendens», «tightening torque» — and it always fails the same way: it swaps them for ordinary words that sound similar.
- It is tuned with material from a field, not from one house: paroxysmal dyspnoea is the same in every hospital. That is why shared tuning is not the cheap option, it is the one that works better.
- The glossary comes first, and it is almost always enough. This is the step above, for whoever already tried it and came up short.
- It is not switched on with a checkbox: you talk, you get a quote, you sign. And it switches off with a button, no questions, immediately.