Transcriptinator — Cherry Blossom Development

Transcriptinator is an on-device voice note app that turns speech into structured notes. It runs Whisper transcription locally, with no internet connection required, and can learn to recognize individual speakers so recordings come out labeled with who said what.

The app also accepts imported audio in common formats (WAV, M4A, MP3, AIFF, ALAC), so existing voice memos and meeting recordings can be transcribed too. Everything happens on the device by default — there's no account, no tracking, and no subscription required for transcription.

Visit Transcriptinator →

Platform Availability

Android

Transcriptinator is available now on Android, distributed through Google Play. It brings the full on-device transcription and speaker recognition experience to Android devices.

Available now on Google Play
Fully on-device transcription
No account required

iPhone

Transcriptinator is available now on iPhone, distributed through the App Store, for iOS 15 and newer. It offers the same local transcription and speaker recognition features as the Android app.

Available now on the App Store
Requires iOS 15+
Feature parity with Android

Features

Local, on-device speech-to-text powered by Whisper — no internet connection required
Speaker recognition through voice enrollment, so transcripts are labeled by speaker
Import existing audio files in WAV, M4A, MP3, AIFF, and ALAC formats
Three processing tiers — Fast (200 MB), Balanced (600 MB), and Accurate (1.4 GB) — to match your device and needs
Custom extraction rules written in plain language, for automatic todo and calendar event creation with zero AI required
Optional AI enhancement using your own API key from Anthropic, OpenAI, or Google for smarter summaries, classification, and date/assignee extraction

Privacy

Audio and speaker data stay on your device. Transcription doesn’t require an account, tracking, or a subscription. If you choose to turn on optional AI extraction, only the text transcript is sent to the AI provider you’ve configured — never the audio itself.