Privacy Policy
For the Voice Transcriber Lite Android application (io.methodox.voicetranscriber.mobile) · Last updated 7 September 2026
Voice Transcriber Lite does not collect, store, or share any information about you. It requests one Android permission — the microphone — and it does not hold the INTERNET permission, so it cannot open a network connection at all. Your speech is transcribed on your own phone, by a model inside the app, and there is nowhere for it to be sent. There are no accounts, no analytics, no advertising, no tracking, and no third-party data-collection SDKs.
What the app is
Voice Transcriber Lite records speech, or opens an audio file you already have, and turns it into text. Speech recognition is performed by OpenAI's Whisper model — the ggml-base-q5_1 build, about 57 MB — which is bundled inside the app package and runs on your phone's own processor. There is no server component and no online service behind the app.
Information we collect
None. The app has no mechanism to collect anything. Specifically, it does not collect, generate, or transmit:
- Names, email addresses, phone numbers, or any other contact details
- Accounts or login credentials — the app has none, and asks you to enter nothing
- Location, contacts, photos, camera, or the contents of your storage
- Advertising identifiers or device identifiers of any kind
- Usage analytics, telemetry, crash reports, or diagnostics
- Your recordings, the audio files you open, or the transcripts produced from them
The microphone permission
The app declares exactly one permission, RECORD_AUDIO, and it is used for one thing: capturing the audio you explicitly ask it to record, when you tap the record button. The microphone is opened when recording starts and released when it stops. The app does not listen in the background, does not record without being asked, and has no wake word or always-on mode.
Android also stops recording when the app is no longer in the foreground — the system silences the microphone for a background app. Rather than pretend to continue, the app stops and transcribes what it captured up to that point.
Audio never leaves your device
Audio you record, and any file you open, is decoded and fed to the bundled Whisper model on your phone. Nothing is uploaded, streamed, or handed to a transcription service, because there is no such service and no way to reach one.
That last point is worth stating precisely: the app does not declare the INTERNET permission. Its AndroidManifest.xml contains a single <uses-permission> entry, for RECORD_AUDIO. Android will therefore refuse the app a network connection. It cannot upload, report, phone home, or check for anything online. This is enforced by the operating system, not merely promised by us.
Files you open
When you open an existing recording — MP3, M4A/AAC, WAV, FLAC, OGG Vorbis or Opus — you choose it through Android's own document picker. The app receives access to the single file you selected and nothing else, which is why it requests no storage permission and cannot browse, list, or read anything else on your device.
Transcripts and the clipboard
A transcript is held in the app while the app is open. The app does not write transcripts to storage; when the app closes, the text is gone.
The one way text leaves the app is the Copy button, which places the transcript on the Android clipboard at your request. From that point the text is in the operating system's hands and goes wherever you paste it. The app never writes to the clipboard on its own, and never reads from it.
Data stored on your device
The app keeps no settings, no history, no saved transcripts, and no library of past recordings. The only file it reads from its own storage is the bundled speech model, which is part of the installed app package.
Because nothing is written, there is nothing to clear and nothing left behind: uninstalling the app removes it completely.
Data retention
None, in the sense the question usually means: we hold no data about you, on any server, for any length of time, because none is ever collected and there is no server. On your device, a transcript lasts only as long as the app is open.
Third parties
There are none. No advertising network, no analytics provider, no crash-reporting service, no cloud backend, and no third-party SDK that collects data is present in the app. No information about you is shared or sold, because none is ever obtained. The Whisper model is an open speech-recognition model bundled into the app as a file; using it involves no communication with anyone.
Children
Voice Transcriber Lite is a general-purpose utility. It is not directed at children, and because it collects no information from anyone, it collects none from children — nothing is knowingly gathered from a child under 13, or under the age of digital consent in your jurisdiction, because nothing is gathered from any user at all. The app contains no advertising, no in-app purchases, and no way to communicate with anyone through it.
Changes to this policy
If the app's behaviour ever changes in a way that affects your privacy — for example, if a future version were to request a permission it does not request today — this page will be updated and the date above revised before that change ships. Material changes will also be noted in the app's release notes.
Contact
Questions about this policy, or about privacy in our apps generally:
Methodox Technologies, Inc.
contact@methodox.io