WhisperBoard is an open-source iOS application designed to provide accessible, high-quality voice transcription on mobile devices. It leverages OpenAI's Whisper model to transcribe audio recordings, making it suitable for users needing to capture thoughts, meetings, or conversations accurately.

How It Works

The app utilizes OpenAI's Whisper model for its core transcription capabilities, offering high accuracy. Users can record audio directly within the app, import existing audio files, and export transcriptions. A key feature is the ability to select preferred microphones and download various Whisper models directly from within the app, allowing for customization of the transcription experience.

Quick Start & Requirements

• Install via cloning the repository and running make.
• Requires Xcode for building.
• Dependencies include OpenAI Whisper and The Composable Architecture.

Highlighted Details

• Powered by OpenAI's Whisper model for accurate transcription.
• Supports recording, importing, and exporting audio files.
• Allows in-app selection of microphones and download of Whisper models.
• Future roadmap includes resumable and real-time transcription.

Maintenance & Community

The project links to OpenAI Whisper and The Composable Architecture. Further community or maintenance details are not provided in the README.

Licensing & Compatibility

Licensed under GPL-3.0. Fonts (Poppins, Karla) are under SIL Open Font License. The GPL-3.0 license may impose copyleft restrictions on derivative works, potentially impacting commercial or closed-source integration.

Limitations & Caveats

The app is currently iOS-exclusive. Real-time transcription is a planned feature, indicating it is not yet implemented. The GPL-3.0 license may require derived works to also be open-sourced.