Mermer collects nothing.

Who this is from

Mermer is developed and published by Richard Nelson, an independent developer. "Mermer", "we" and "the app" below all mean the same thing: the Mermer app on your iPhone.

What Mermer accesses on your device

Mermer asks for permission to use the microphone. iOS shows you that request, and you can withdraw it at any time in Settings → Privacy & Security → Microphone.

Mermer records only after you start it, and it can continue recording while the screen is locked until you stop it. iOS shows its own recording indicator throughout. Mermer does not start recordings autonomously, and it has no wake word.

What Mermer stores, and where

Everything Mermer keeps is stored locally on your device, inside the app's own private container, and is included in your device backups exactly as any other app's data is.

• Your recordings — the audio files themselves, as ordinary .m4a files.
• Transcripts — the text of each recording, with timings, saved alongside the audio. If you edit a paragraph's text or correct who said it, the original is kept too, so it can be restored.
• Speaker names — any name you type in place of "Speaker 1", stored with that recording.
• Remembered voices — see the section below.
• Settings — whether you like timestamps shown, and your chosen transcription language.
• A diagnostics log — a small local file recording timestamps of things like audio session events, kept for troubleshooting. It never contains audio or transcript text and is never uploaded; see "Diagnostics" below.

None of this is uploaded anywhere by Mermer. Deleting a recording removes its audio and transcript from Mermer: an Undo option stays available for a few seconds, and the recording is cleared from Mermer's trash the next time you open the app. Copies you have shared, and copies already included in a device backup, are not affected by that deletion. Deleting the app removes everything Mermer stored.

Remembered voices

Mermer can optionally remember a voice so that it labels that person by name in future recordings. This is the most sensitive thing the app does, so it is worth being precise about it.

• It only happens when you ask. No voice is ever remembered automatically. You tap a speaker, give them a name, and explicitly choose to remember that voice.
• What is stored is not a recording. The profile stores numerical voice characteristics, not a playable recording, derived from the audio the way a fingerprint hash is derived from a fingerprint. Recognition based on it is approximate, not certain, so it's worth checking a name before you rely on it.
• It stays on your device, in the app's private container, and is never uploaded, shared or synced by us.
• You can delete it. The Remembered Voices screen lists every voice you have saved and lets you rename or delete any of them. Deleting a profile removes it from your device, and Mermer stops recognising that voice going forward. Names already saved in past transcripts are not changed, and the recordings that contain that person's voice are separate and are not affected.

Because a voice belongs to a person, please get their agreement before you remember their voice, or before you record them at all. See "Recording other people" below.

Transcription happens on your iPhone

Mermer transcribes using Apple's on-device speech model, part of iOS itself. Your audio is handed to that model on the device and the text comes back on the device. It is not sent to Apple, to us, or to anyone else for transcription. Transcription depends on your device supporting a language covered by Apple's model; recording still works without it, and processing time varies with the length of the recording. You can choose a transcription language under About Mermer → Transcription Language, or for a single recording under More → Language.

Speaker separation works the same way: the models that tell voices apart are included inside the app, so that step needs no network access either.

Network activity

Mermer does not upload recordings, transcripts or speaker profiles. There is no developer-operated server for it to send them to, and no advertising or analytics SDKs that would phone home on their own.

One kind of network activity can still occur. A model download may be needed for on-device transcription: iOS manages speech model installation and updates, not Mermer, the same way it does for every app that uses the model. That download goes from Apple to your device, and nothing is uploaded in the process: no audio, no text, and nothing about you.

Diagnostics

Mermer writes a small local troubleshooting log inside its own container: timestamps of things like audio session events, kept so a bug report can be tracked down. It never contains audio or transcript text, it is never uploaded or sent anywhere, and it is deleted when you delete the app. No crash-reporting or analytics service is integrated into Mermer.

Recording other people

Mermer is a tool for recording conversations, and the laws about recording other people differ from place to place. Whether and how you may record someone is your responsibility as the person doing the recording. We would add the obvious non-legal point too: tell people they are being recorded.

What Mermer does not do

• No accounts, sign-in, or user profiles.
• No servers or backends operated by us: there are none.
• No analytics, telemetry, or usage tracking of any kind.
• No crash-reporting or analytics service is integrated.
• No advertising or analytics SDKs. Speaker separation uses a bundled, open-source, on-device library (FluidAudio), with its models shipped inside the app rather than fetched from anywhere.
• No advertising, and no advertising identifier.
• No cloud transcription, and no AI service anywhere off the device.
• No selling or sharing of data, because there is no data to sell.
• No tracking of you across other apps or websites.
• No in-app purchases or payment processing.

Mermer's App Store privacy label is "Data Not Collected". Mermer does keep a small local diagnostics log for troubleshooting, covered under "Diagnostics" above; it is never uploaded and does not count as collection.

Sharing is yours to initiate

Mermer can share a recording or a transcript through the standard iOS share sheet. That happens only when you choose it, and where it goes — Messages, Mail, Files, another app — is your choice. Once you have sent something somewhere else, that destination's own privacy policy applies to it, not this one.

Children

Mermer is rated 4+. Because it collects no information from anyone, it collects no information from children either.

Your rights

Data-protection laws such as the GDPR and the CCPA give you rights to access, correct, export and delete the personal data a company holds about you. We hold none, so there is nothing for us to produce or erase.

The data Mermer keeps on your own device is already entirely under your control: delete a recording, delete a remembered voice, or delete the app to remove everything at once.

Changes to this policy

If Mermer's behaviour ever changes in a way that affects this policy, this page will be updated and the effective date at the top will change. Since the app has no analytics, the honest way to see what changed is to read this page — it is deliberately short enough to re-read.

Contact