Whisperboard is an open-source iOS app that brings OpenAI’s Whisper speech recognition to the iPhone for high-quality voice transcription, using whisper.cpp so audio can be transcribed privately on the device. Released under the GPL-3.0 license, it has about 1,100 stars on GitHub and was last updated in 2025.
Recording a thought and getting accurate text back should not require uploading your voice to a cloud service. Whisperboard is built around that idea: tap once to record, and the app transcribes with Whisper models that you can browse and download in different sizes to trade speed against accuracy. You can import existing audio files for transcription, export recordings and transcripts for sharing or editing, choose which microphone to capture from, and resume interrupted transcriptions where they left off. The app is written in SwiftUI using The Composable Architecture, which makes it a clean, well-structured codebase as well as a practical tool. For journalists, students and anyone who thinks out loud, it turns an iPhone into a private transcription pocket recorder powered by one of the strongest open speech models available.
Whisperboard is free and open source under the GPL-3.0 license. The source code is on GitHub for anyone to build and use; transcription runs on your device, so there are no per-minute cloud fees.
Whisperboard is for iPhone users who record meetings, lectures, interviews or voice notes and want accurate, private transcription without a subscription service, and for iOS developers studying a well-architected Whisper integration.