Signing in & your settings
Welcome to minutes — your self-hosted, EU-friendly tool for live meeting transcription and translation. This page gets you from zero to signed in, and walks you through your personal settings.
What you need
Before you start, gather a few things:
- A minutes account. There's no public signup — your administrator creates your account for you and gives you an email and a temporary password.
- Your server's URL. This is the web address your team runs minutes on, something like
https://minutes.your-org.com. Your admin will tell you this too. - Something to capture. minutes doesn't dial into calls. You can capture in three ways: join a Google Meet/Microsoft Teams call (or open any tab that plays audio — a webinar, a video, a podcast) and let the browser extension capture that tab's audio; record your own microphone right in the app (on a computer or a phone, no extension); or upload a file you already have.
- Your own API keys:
- A Soniox key — get one at soniox.com. Powers transcription — both live capture and uploaded files.
- An Anthropic key — get one at console.anthropic.com. Needed for translation.
Transcription uses your Soniox key, and translation uses your Anthropic key — set both under Settings → API keys. Your administrator may configure a shared Soniox key that live capture falls back to if you haven't added your own, but the recommended setup is your own key, with the region (US/EU) that matches it. More on this below.
Signing in
- Open your server's app in a browser. The app lives at the
/apppath — for example,https://minutes.your-org.com/app. (On a plain self-hosted setup with no marketing landing page, the app may be at the root,https://minutes.your-org.com/.) - Enter the email and password your administrator gave you, and click Sign in.
That's it — you'll land on the main view, with your list of Transcriptions on the left and the transcript and translation columns to the right. (On a phone, the same app shows one screen at a time — see On your phone below.)
If you don't have an account yet, ask your administrator — there's no self-service signup. The sign-in screen says the same thing.
Change your password
Your admin set your first password, so change it to something only you know.
- Click your email in the top-right corner, then choose Settings.
- On the Account tab, find Password.
- Enter your current password, then a new password, and click Update password.
Your new password must be at least 12 characters and use at least 3 character classes (for example, lowercase, uppercase, and digits). The form will tell you if it's too weak.
Changing your password signs out every other session — including the browser extension and any other browser — but keeps you signed in where you made the change. You'll need to sign in again from the extension afterward.
A tour of your settings
Open Settings from the menu under your email (top-right). There are four tabs: Account, API keys, Translation, and Danger zone.
API keys
This tab is where you paste your own provider keys, stored encrypted — once saved, a key is never shown back to you (you'll see a "set" marker instead, and can replace it at any time).
- Soniox API key — powers speech-to-text for both your live captures and your audio uploads. Paste your key and click Save. (If your admin configured a shared server key, live capture falls back to it when you haven't set your own — but uploads always use yours.)
- Soniox region — United States or European Union. This is data residency: pick EU to have your audio processed on Soniox's EU endpoints. It must match the region of the Soniox project your key was created in — an EU key only works with EU selected. Your choice saves immediately and applies to both live capture and uploads.
- Anthropic API key — powers translation (live, on uploads, and the on-demand "translate this line" button). Without it, translation is simply off. Paste your key and click Save.
Here's the full picture of which key does what:
| Feature | Whose key is used |
|---|---|
| Live capture transcription | Your Soniox key + region (falls back to the server key, if your admin set one) |
| Audio-upload transcription | Your Soniox key + region |
| Translation (live, upload, on-demand) | Your Anthropic key |
If you (or your customers) need audio to stay in the EU, create your Soniox key in an EU project and select European Union as your region. See Configuration → Data residency for the operator's side.
Translation
These are your defaults — applied automatically when a new meeting starts. You can always override them for an individual meeting from the translation column's gear icon.
- Translate new meetings by default — a toggle. When on, translation starts automatically for each new meeting.
- Default output language — the language your transcripts get translated into. Choose from English (en), German (de), or Persian (fa), or leave it blank for none.
- Default model — which Claude model translates: Haiku (fastest, cheapest), Sonnet (balanced), or Opus (best quality). Leave it on the default if you're unsure.
Click Save defaults when you're done. You can override the language and model per meeting from the translation column's gear icon (⚙).
Translation needs your Anthropic key (see the API keys tab). If you turn defaults on but haven't set a key, translation stays off until you add one.
Account
This tab shows your email (and an "admin" marker if you're an administrator) and is where you change your password, as described above.
Danger zone
The last tab holds Delete account — a permanent, self-service way to remove your account. You'll be asked to type your email to confirm. Meetings you own become unowned (an admin can still see them), and the action can't be undone.
Two audio sources in one meeting
A single meeting can capture two independent audio sources at once, kept fully separate — each keeps its own transcript, its own translations, and its own timestamps:
- Online stream — the audio coming from the browser tab you're capturing (the Meet/Teams call, a webinar, a video). Marked in orange.
- Host mic — your own microphone, the person doing the capturing. Marked in blue.
(A third, single-source kind, Upload, is what you get from an uploaded file — it never has a second source.)
When a meeting has both sources, a small segmented control appears in the meeting header — Online stream | Host mic, each with its colored mark. Click either one to switch which source you're reading: the transcript and its translation switch together, instantly, with no reload. A meeting that only ever had one source shows no switcher — there's nothing to choose between.
The two sources stay separate — you read one at a time. A combined, side-by-side view of both at once isn't available yet; it's a planned addition.
Switching sources only changes what you're looking at — both sources keep recording and translating in the background. Switch back and the other source's lines are right where you left them.
Copy the transcript or translation
Need to drop a transcript into an email, a chat, or your notes? Copy grabs the whole thing as plain text — no download, no timestamps, just the words ready to paste.
- On a computer, a small copy icon sits in the Transcript column header and another in the Translation column header.
- On a phone, the copy icon is in the meeting header; it copies whichever of Transcript / Translation you're currently viewing (the toggle at the top).
What gets copied is the source you're viewing — copy handles each source separately. If a meeting has both audio sources, the copy reflects whichever one the source switcher is set to; switch to the other and copy again to grab it. Transcript and translation are likewise copied separately — the Transcript copy gives you the original lines, the Translation copy gives you the translated lines.
Copy is the fast grab: the visible source's lines, no timestamps, straight to your clipboard. When you want a file — txt / md / json, with the choice of transcript / translation / both, timestamps on or off, and (for a two-source meeting) which source — use Export instead.
A copy only has translated text for lines that were actually translated. Untranslated lines are skipped on the translation side, so translate the lines you want before copying the translation.
On your phone
minutes works in a mobile browser too. On a small screen the three columns collapse into one view at a time:
- Your Transcriptions list is the home screen — tap a row to open it, or the ⟳ to refresh.
- A meeting opens full-screen with a Transcript / Translation toggle at the top; tap ‹ to return to the list.
- When a meeting has both audio sources, a row of source chips (Online stream / Host mic, each with its colored mark) sits just above that Transcript / Translation toggle — tap a chip to switch which source you're reading, the same as the desktop switcher. Single-source meetings show no chips.
- The account menu, a meeting's actions (rename, export, share, delete), and the per-meeting translation settings each slide up as a bottom sheet.
You can also record your own microphone right on your phone — tap Record on the Transcriptions screen for a full-screen capture flow. (The capture extension, which grabs a browser tab's audio, is desktop-Chrome only.) And of course you can read, translate, copy, share, and export from your phone. For the smoothest experience, install minutes to your home screen so it runs full-screen like a native app.
Where to next
- Set up the browser extension — install it, point it at your server, and start capturing meetings.
- Record in the app — capture your own microphone right in the web app (computer or phone), and install minutes to your home screen.
- Meetings & export — rename meetings, translate lines, copy, share, and export transcripts.
- Uploads — transcribe an existing audio or video file (this is where your Soniox key comes in).