Your first meeting with Akume

Akume joins your call the way a colleague does: you paste the meeting link, the translator shows up in the participant list, and from that moment every speaker is translated as they talk. The translator speaks the translation aloud right in the call, captions appear on its tile, and anyone can also follow along on a listener link in their browser. Nobody installs anything.

Before you start

• A meeting on Google Meet, Microsoft Teams, or Zoom, and its join link.
• Someone in the call who can admit a participant, in case a waiting room is switched on.
• An Akume account with a plan. Before the first meeting the app asks you to choose a plan and enter a card, and the plan or intro is charged at checkout, so do this before the call rather than in its first minute.
• A sense of how long your minutes last. The language being spoken is not counted, so a minute of the call uses one minute fewer than the number of languages selected: 2 languages = 1 minute, 3 languages = 2. So each intro minute covers about one minute of a two-language meeting: the details are in Free demo and paid intro.

Run the meeting

1. Create your account. Sign up at app.akume.ai and choose a plan; an intro is the cheapest way to start.
2. Copy the meeting link. Take it from the calendar invite: a meet.google.com, teams.microsoft.com (or teams.live.com), or zoom.us address. Copy it whole: on some platforms the part after the question mark carries the passcode, and a trimmed link cannot be joined.
3. Paste the link into Akume and list the languages spoken in the room. It is one list, not a “from” and a “to”: you name every language people will speak, from 2 to 10 of them, and Akume works out who is speaking which one as they talk. Two regional variants of the same language, such as English (US) and English (UK), count as one, so the list still needs two genuinely different languages.
4. Send the translator in. Akume dials into the meeting as a participant. If a waiting room or lobby is enabled it waits there, and the host admits it with the same click any other guest gets.
5. Share the listener link. It is optional: everyone in the call already hears the translation. The link adds captions and a language of their choice on their own screen. Each meeting has its own. As soon as the translator is in the call it posts the link in the meeting chat itself (a short message in English, pinned in Google Meet and Microsoft Teams), and Akume also copies it to your clipboard the moment the session starts, so you can paste it yourself if the chat message is missed. Once the session is Active you can copy it again from the session screen. Whoever opens it in a browser chooses a language; if the status line asks them to press play, one tap on Play starts the audio, and the same button pauses. From there they follow the call with captions and translated audio, with no account and nothing to install. A single session allows up to 500 listener connections this way. Following a meeting with the listener link is written for the people you send it to, so paste it alongside the link if they have not used Akume before.
6. Talk normally. Everyone speaks their own language, and the translator voices each turn in the other languages right in the call while the conversation continues. People who prefer to read simply read the captions instead.
7. Finish. When the call ends the translator leaves with everyone else, and you can remove it earlier from your dashboard. Minutes are counted only while it is in the call, and the language being spoken is not counted: 2 languages use 1 minute per minute, 3 languages 2. Keep the listener link if you want to reread the captions later. Privacy and data protection explains how.

If something goes wrong

The translator never appeared

• Check the waiting room first: most first joins are simply waiting to be let in.
• Confirm the link is the join link for the meeting itself, not a calendar page or a shortened link that lost its passcode.
• Start the meeting before sending Akume in; an unstarted call has nobody to admit it.

The translator left in the middle of the call

Almost always the minute balance reached zero. After that the session runs on a grace allowance of 10 billed minutes (about ten more minutes of call with two languages, five with three), and then the translator leaves and the session stops. Starting a fresh one needs at least 5 minutes of the call on the balance (5 minutes with two languages, 10 with three), so an almost-empty account cannot simply begin again. Keep the rule in mind when you estimate: with three languages each minute of the call uses two minutes.

A long call stopped after eight hours

A single session is capped at 8 hours. Start a new session for the rest of the meeting and share the new listener link: the previous link belongs to the session that has ended.

A listener says they hear nothing

Two ordinary explanations. Either the audio is paused (Akume starts it as soon as a language is chosen, but a browser can refuse that first attempt, and the status line on the page then asks for a tap on Play), or nobody has spoken a different language yet. Spoken translation is produced only when somebody speaks one of the other languages: nobody is served a translation of their own. The captions do not go quiet in the meantime. Every utterance is captioned, and the ones in the listener’s own language appear as the original text.

Somebody’s language is missing from the session

The list is fixed when the session starts and cannot be edited mid-meeting. Stop the session, start a new one with the full list of languages, and share the new listener link.

The platform-specific details have their own guides: Using Akume with Google Meet, Using Akume with Microsoft Teams, and Using Akume with Zoom.