Kaseta

Offline transcription that never uploads your audio

Offline transcription means the speech recognition runs on the device that holds the audio. Not “encrypted in transit”. Not “deleted after 30 days”. The recording is never sent, because the work happens where it already is. Kaseta transcribes 99 languages this way on an iPhone, and the difference from a cloud service is not a promise — it is a network request that does not exist.

Three things offline has to mean

It works with the radios off. There is no account, because an account implies a server that knows you. And nothing is queued for a quiet upload later, once you are back on Wi-Fi. Any one of those missing and “offline” is a mode, not an architecture. Kaseta has no servers at all, which is a stronger statement than a promise not to use them: there is nowhere for the audio to go.

What is actually running on the phone

A speech recognition model, shipped inside the app, executed by the phone’s own hardware. The free plan uses a smaller, faster one; Pro adds a quality mode built on a larger model for difficult audio. Both live on the device. Neither of them phones home, which is also why the app has nothing useful to log about you.

What it costs you to do it this way

Honest ledger. Long files take real time and real battery, and a five-year-old iPhone will be slower at it than a current one. There is no web dashboard, no team workspace, and no bot that joins your meetings. Speaker labels are not there yet. And the free plan transcribes 5 minutes of each recording, though the recording itself is never cut short — audio keeps going and waits.

When a cloud service is the better tool

Why there is no subscription

The costs of a cloud transcription service scale with your minutes, so it charges you by the minute. Ours do not exist, so we do not. Pro is €19.99 a year or €39.99 once, and the once-priced version is the point rather than an upsell.

Next: how to transcribe a voice memo is the practical version of this page, and the homepage has the short version of the same argument.

Steps

  1. 01

    Put the phone in airplane mode

    Open Control Centre and switch the radios off, Wi-Fi included. The point is to leave the app with nowhere to send anything, not to trust that it would not.

  2. 02

    Record something

    Press record and talk. Recording starts immediately, because there is no session to open with a server first.

  3. 03

    Watch the text appear anyway

    The transcript is computed by the phone as you speak. If the work were happening in a data centre, this is the step that would fail.

  4. 04

    Turn the network back on

    Nothing uploads late, because nothing was queued and there is no account for it to sync to. The test you just ran is the whole architecture.

Frequently asked questions

Can I transcribe audio offline?

Yes. Offline transcription means the speech recognition model runs on the device holding the audio instead of on a remote server, so it works in airplane mode, on a plane, or in a building with no signal. Kaseta does this on iPhone; on a desktop, a local Whisper build does the same thing.

What is on-device transcription?

It is speech recognition performed by the phone's own processor, using a model shipped inside the app, with no network request in the loop. The practical difference from a cloud service is not the wording of a privacy policy — it is that the upload never happens, which is something you can verify yourself by cutting the network.

Is offline transcription as accurate as the cloud?

Usually close, sometimes behind. Cloud services can run very large models on server hardware, which shows on hard audio: heavy accents, crosstalk, a noisy room. Kaseta ships a smaller model on the free plan and a higher-accuracy mode on Pro, and on clear speech recorded near the phone the gap is small enough not to matter.

Do I need an account to use it?

No. There are no accounts, because there is no server to hold one. Nothing to sign up for, no email address to hand over, and no cookies on this website either.

Related guides