The private voice recorder for iPhone.
Miru listens, transcribes directly on the device and automatically separates speakers. Nothing is sent to a cloud for recognition. You share the finished transcript however you like.
Apple's new SpeechAnalyzer (iOS 26) runs entirely offline on your iPhone. Once the language model is downloaded, no audio leaves the device.
Miru automatically clusters by voice during the recording. After stop you confirm or correct the assignment — you can also split clusters manually.
Markdown straight to the clipboard for ChatGPT. Or share as TXT, Markdown file or PDF — via the standard iOS share sheet.
No ad IDs, no analytics, no third-party trackers. Sync optionally runs through your own account to my private Supabase instance.
Big button in the middle. Miru keeps running in the background when you put your iPhone away.
After stop a sheet opens with the detected voice clusters. Per cluster: assign known person, create new speaker, save anonymously or discard.
In the detail view you tap the share icon top right. Copy markdown, or as a file (TXT/MD/PDF) to Mail, Notes, ChatGPT, whatever.
Miru is currently in closed beta. If you want to test it, drop me a line and I'll invite you on TestFlight.