Offline audio/video transcription desktop app using OpenAI Whisper with privacy-focused local processing.
Vibe is a desktop application that provides offline transcription of audio and video files using OpenAI's Whisper model. It prioritizes user privacy by processing all data locally on the device, ensuring no sensitive information is sent to external servers. The application supports batch processing, multiple export formats, and advanced features like speaker diarization and translation.
Vibe is designed for content creators, researchers, journalists, and developers who need to transcribe audio or video files offline with a strong emphasis on data privacy and local processing. It also caters to users requiring batch processing, multi-format export, or integration with platforms like YouTube.
Developers choose Vibe for its fully offline, privacy-first approach, eliminating data transmission to external servers. Its unique selling points include extensive platform support, GPU acceleration across major operating systems, and advanced AI features like local summarization via Ollama and a built-in HTTP API for integration.
Transcribe on your own!
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
All transcription is processed locally on the device; no data is sent to external servers, as emphasized in the README's privacy-first approach.
Supports SRT, VTT, TXT, HTML, PDF, JSON, and DOCX exports, catering to diverse needs like subtitling or document creation.
GPU acceleration for Nvidia, AMD, and Intel GPUs on macOS, Windows, and Linux ensures faster processing, as highlighted in the features.
Includes speaker diarization, translation, and local summarization via Ollama, enhancing transcription utility beyond basic text conversion.
iOS and Android versions are listed as 'coming soon' in the README, limiting current use to desktop platforms only.
Without a capable GPU, transcription can be slow, especially in stable timestamps mode which the README notes is slower.
Summarization via Claude API needs an internet connection, partially contradicting the offline privacy claim, though Ollama offers a local alternative.