Local-first AI subtitle generator that runs offline and plugs into DaVinci Resolve, Premiere Pro and After Effects.
4.4K305TypeScriptMITUpdated 15 days ago
- Stars
- 7,837
- Forks
- 567
- Open issues
- 234
- Last commit
- 14 days ago
What is it?
Vibe is a cross-platform desktop application for macOS, Windows and Linux that converts speech in audio and video files into text. All transcription runs on your own machine, so recordings never leave the device.
The app supports Whisper, Nemotron and Parakeet TDT models and can export transcripts as SRT, VTT, TXT, HTML, PDF, JSON or DOCX. It also offers batch processing, realtime preview, speaker diarization, translation to English, and transcription from the microphone, system audio or a video URL from popular sites.
For power users, Vibe includes a command-line interface and an HTTP API with Swagger docs. Optional summarization works through the Claude API or local models via Ollama, and a phone can be used as a recording source by scanning a QR code.
Key features
- 01Offline transcription of audio and video with no data uploaded
- 02Supports Whisper, Nemotron and Parakeet TDT models
- 03Exports to SRT, VTT, TXT, HTML, PDF, JSON and DOCX
- 04Batch transcription, realtime preview and speaker diarization
- 05CLI and HTTP API with Swagger docs
- 06GPU acceleration via Vulkan and CoreML
Use cases
- Generating subtitles for videos or reels
- Transcribing interviews, lectures or meetings without cloud services
- Converting recordings from a video site into searchable text
- Translating non-English audio into English text
Quick start
- 1
Download the app from the project's download page
- 2
Read the local development guide to build from source
Things to know
- Local development setup is documented separately in docs/BUILDING.md, so building from source takes extra effort.
- Pull requests are limited to a few collaborators, so outside code contributions are restricted.
Topics
FAQ
- What is vibe?
- Desktop app that transcribes audio and video fully offline on your own device, using Whisper and other speech models. Vibe is a cross-platform desktop application for macOS, Windows and Linux that converts speech in audio and video files into text. All transcription runs on your own machine, so recordings never leave the device.
- Is vibe free to use? What license does it have?
- vibe is open source under the MIT license. Check the license terms before using it in commercial products.
- Is vibe still maintained?
- Actively maintained. The last commit to vibe was on Oct 6, 2026.
- What can I use vibe for?
- Generating subtitles for videos or reels Transcribing interviews, lectures or meetings without cloud services Converting recordings from a video site into searchable text Translating non-English audio into English text
- How do I install vibe?
- Download the app from the project's download page Read the local development guide to build from source
- Where can I get vibe?
- The source code and installation instructions are on GitHub: https://github.com/thewh1teagle/vibe
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Desktop app that transcribes and translates audio and video offline on your computer, built on OpenAI's Whisper.
22.1K1.6KPythonMITUpdated 3 days ago
C/C++ port of OpenAI's Whisper speech recognition model for fast, offline, dependency-free speech-to-text on CPU and GPU.
54.5K6.3KC++MITUpdated 24 hours ago
Open-source, local-first AI dictation app that turns speech into text on macOS and Windows, working offline with Whisper.
1.6K142TypeScriptMITUpdated 7 days ago
A faster reimplementation of OpenAI's Whisper speech-to-text model on the CTranslate2 inference engine, with lower memory use and 8-bit quantization.
26K2.2KPythonMITUpdated 4 days ago
Open-source speech-to-text built on Whisper that adds word-level timestamps via forced alignment and optional speaker diarization.
24.7K2.5KPythonBSD-2-ClauseUpdated 2 days ago