whisper-ui

• Purpose: Whiskey is a GUI for OpenAI's Whisper, designed to transcribe audio/video files into text using Tauri and SvelteKit. It leverages whisper.cpp for Whisper's C++ binaries.
• Key Features:
  • Transcribes audio/video files to text.
  • Real-time text highlighting during playback.
  • Supports exporting transcripts as .txt or .vtt files.
  • Includes drag-and-drop functionality and video file support.
• Planned Features:
  • Mic recording, multi-platform binaries (Apple Silicon, Linux, Windows), editable text, and UI error displays.
• Framework: Built with SvelteKit (frontend) and Tauri (backend).
• Target Audience: Users needing a desktop GUI for Whisper's transcription capabilities.

Links:

• Whisper
• whisper.cpp