A voice recorder with built-in speech-to-text. Everything runs on your iPhone: audio never leaves the device, works offline. Speakers are labeled automatically.
Letopis is a private voice recorder with on-device speech recognition. Meetings, interviews, lectures, and voice notes become a verbatim transcript right on your iPhone: no cloud, no accounts, no analytics. Audio and text never leave the device.
It is built for conversations whose content must not leave the phone: deal negotiations, legal consultations, medical and counseling sessions, journalist and researcher interviews, hiring conversations. Letopis sends neither sound nor text to any server: speech recognition and speaker labeling run on the iPhone's processor, with no internet involved.
The transcript is verbatim. Every line carries its speaker and timecode; the app writes down what was said and adds no summaries or conclusions. For work where the exact wording matters.
Recording starts with one tap; a level meter and timer stay on screen. When you stop, Letopis transcribes the recording and separates the speakers.
Share audio or video from any other app and Letopis transcribes it the same way: messenger voice notes, including Ogg/Opus, dictaphone files, video clips. A video keeps only its audio track. Duplicates are detected by content, not by file name.
What else the app does:
- A built-in transcript editor: fix lines, rename and merge speakers.
- Personal-data masking in the text: names, organizations, addresses, numbers - also on-device.
- A synced player: tap a timecode to hear that exact moment of the recording.
- Search across names and transcript text; recordings also appear in iOS search.
- Custom recording names, export of text and audio.
- Two recognition engines: GigaAM v3 for Russian and multilingual Parakeet (25 languages).
- An optional reminder to ask participants for consent.
Everything stays on your iPhone. The only network access is a one-time model download on first launch. No accounts, no ads, no analytics, no subscriptions; App Privacy says "Data Not Collected".
Requires iOS 26 or newer.
Ratings & Reviews
This app hasn’t received enough ratings or reviews to display an overview.
Ready for iOS 27. Transcription in the background keeps its Neural Engine access: iOS 27 closes it to background work without a separate permission, and the app now holds that permission.
The speaker separation engine is updated: same result, a little faster.
Delayed and repeated transcriptions now save their actual completion time. Repeated recording stops are handled correctly.
Version 0.1.482
The developer, Sergey Makarov, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy .
Data Not Collected
The developer does not collect any data from this app.
Privacy practices may vary, for example, based on the features you use or your age. Learn More
The developer indicated that this app supports the following accessibility features. Learn More
Supported Features
VoiceOver
Voice Control
Larger Text
Dark Interface
Differentiate Without Color Alone
Reduced Motion
Captions
Information
Seller
Sergey Makarov
Size
40.2 MB
Category
Business
Compatibility
Requires iOS 26.0 or later.
iPhone Requires iOS 26.0 or later.
Mac Requires macOS 26.0 or later and a Mac with Apple M1 chip or later.