Real-time call translation — while they are still talking
Whisperer follows the call and puts the translation under the live transcript, in the same window above your screen. No switching tabs, no asking anyone to say that again.
Translation is on the paid plans · macOS and Windows
The answers below are an example. Your own base connects in the app.
…
…
Let's start with Q3 — how did conversion move?
Conversion is up 14% — let me pull up the funnel.
And the cost per lead?
Cost per lead is €12.40 — 9% under plan. Most of the gain came from organic.
Not a video — this is Whisperer itself, and the panel really answers. The microphone starts only when you press it.
EN → RU
Two lanes, moving in step
On the left what is being said, on the right what you read. Press Play: the translation arrives after the line, not alongside it. That is how it works — claiming “no delay” would be untrue.
-
Them · RUПилот, по сути, закончен. Сейчас нам нужно решение по цене.EnglishThe pilot is basically done. What we need now is a decision on pricing.
-
You · ENWe are ready to discuss pricing, but first I want to understand the volume for next quarter.РусскийМы готовы обсуждать цену, но сначала хочу понять объём на следующий квартал.
-
Them · RUПримерно столько же, может быть на пятнадцать процентов больше, если интеграция взлетит.EnglishRoughly the same, maybe fifteen percent more if the integration lands.
A recorded example of the two lanes. The live demo is in the panel above: it listens only when you press.
Limits
Two things to know up front
Both follow from how the product is built, and neither is a setting. The empty lane above is the first one.
-
First
One recording at a time, not an archive migration
A finished recording can be uploaded: the Meetings section takes one file up to 25 MB — mp3, m4a, wav, ogg, flac, aac, plus mp4 and webm, from which the audio track is used. What is missing is bulk upload of a whole archive; and it spends the same minutes a live call does.
-
Second
The other side comes through system audio
Their speech arrives through system audio: on macOS that means granting the Screen Recording permission, on Windows no extra permission is needed. Without it the transcript and the translation contain your lines only — and theirs are the ones you came for.
Setup
Three switches and that is it
Translation lives in the client settings, not in a separate app. No bot in the call, no integration with the conferencing service.
Target language
The language you want to read. The translation line appears under the transcript; the original stays where it was.
Source language
Auto-detected or pinned. If the call runs strictly in one language, pin it — the accuracy is better that way.
Their side only
Useful on bilingual calls: your own language needs no translation, and the extra line only gets in the way.
What this looks like in specific services, and what to do when the other side is not audible, is covered in A live interpreter for Zoom, Google Meet and calls.
Read or hear
There are two kinds of translation, on different plans
The difference is not quality but whether you read the translation or hear it. We say so before you sign up, so the plan is not a surprise afterwards.
Translation as text in the window
A translation line under the live transcript, with both sides translated independently. That is what the example above shows. On the Free plan the translation line does not appear.
Simultaneous translation as voice
A separate mode where the translation is spoken, not only written. It is part of the Max plan and draws on the same weekly premium budget as the frontier models.
Live Translation
Two modes of voice translation
Both translate for real. The trade-off differs: standard waits for a finished thought and sounds more natural, simultaneous starts earlier and builds the translation from a phrase you have not finished yet.
Standard
Waits for the end of your phrase, then translates it whole. Smoother and more natural.
Simultaneous
Translates as you speak: the other side hears you about two seconds in.
Questions
What people ask first
How large is the delay?
The translation appears a second or two after the original line. Both are visible at once: the original in the transcript line, the translation in the line below it. A noticeably longer delay usually means the connection — translation is an extra network round trip.
Which languages does it work with?
Recognition runs on Whisper, which supports more than 90 languages; the session language is set before you start. The target language and the source language are chosen separately in the client settings.
Can I upload a finished recording?
Yes. In the Meetings section you pick the file with a button or drag it in — one at a time, up to 25 MB: mp3, m4a, wav, ogg, flac, aac, plus mp4 and webm, from which the audio is taken. From there it goes exactly like a live call: transcript, meeting map, decisions and tasks. A whole archive cannot be uploaded in one go, and in no-logs mode upload is unavailable — nothing is stored there by design.
Does anything need installing in Zoom or Google Meet?
No. Whisperer is a window above your screen, not a plugin for the conferencing service and not a participant. It runs over Zoom, Google Meet, Microsoft Teams, Telegram, Discord and anything else.
Will the other side know I am reading a translation?
The window is visible to you only, and it stays out of a shared screen when capture exclusion is on. Nothing stops you from saying you are using translation either — on international calls that is unremarkable.
Is translation included in the free plan?
No. Free gives 60 minutes of recognition a month, but the translation line does not appear — it opens on the paid plans. Simultaneous translation as voice is part of the Max plan.
What is left after the call?
The full transcript in your account history, a meeting map with topics, decisions and tasks, and search across what was said. If you would rather keep nothing, switch on no-logs mode.
Try it on one call
Take the next conversation with a colleague abroad, or sit in on any webinar in a language you half-know. Fifteen minutes is enough to tell whether the translation line helps or distracts.