Vibe

Vibe is a desktop app for people who need transcripts or subtitles from recordings without uploading them to a transcription service. It handles audio and video locally, including interviews, meetings and material in different languages.

• Transcription and translation: Use Whisper, Nemotron or Parakeet models to transcribe multilingual recordings and translate speech into English.
• Files and live audio: Batch transcribe files, capture microphone or system audio, or record on an iOS or Android phone while your computer handles transcription and returns the text.
• Subtitles and speaker labels: Distinguish speakers, follow a live transcript preview and adjust caption length for videos or short clips; a slower timing mode targets subtitle accuracy.
• Transcript exports: Save subtitles as SRT or VTT, or export text and documents in TXT, HTML, PDF, JSON and DOCX formats.
• AI analysis: Analyze transcripts and create batch summaries locally with Ollama, or use the optional Claude API for multilingual summaries that send transcript content to a cloud service.

Vibe is open source under the MIT license and runs on macOS, Windows and Linux. Transcription works fully offline. GPU acceleration supports NVIDIA, AMD and Intel hardware through Vulkan or CoreML. Transcribing media from sites such as YouTube and Vimeo requires access to those online sources. A command line interface and HTTP API let other software submit transcription work.