Vibe
An offline transcription app under MIT that records system audio and the microphone, supports Whisper, Parakeet and Nemotron models, labels speakers, and exports SRT, VTT, TXT, HTML, PDF, JSON and DOCX.
What it does
Vibe's one line is "Transcribe audio and video right on your device". It takes files in batches, system audio, the microphone, website links (YouTube, Vimeo, Facebook, Twitter are named) and a phone recording through a QR code. The README lists Whisper, Nemotron 3.5 and Parakeet TDT v3 models, with custom models added through a link scheme.
It labels speakers (the README uses the term speaker diarization), translates any language to English, and exports SRT, VTT, TXT, HTML, PDF, JSON and DOCX. Summaries use either the Claude API, which is a cloud call, or a local model through Ollama. There is a command line, an HTTP API with Swagger docs, a realtime preview, and a stable-timestamps mode the README says is slower.
What it replaces
Vibe replaces Otter.ai and Fireflies.ai for recordings and for calls captured from your own machine's system audio, with the widest export list on this site. Its README puts privacy first: "fully offline transcription, no data ever leaves your device".
Where it falls short: it does not join calls as a bot. The summary path through Claude sends the transcript to Anthropic; the Ollama path keeps it local. Push-to-talk dictation into other apps is not part of it.
How it runs
Download from the project site, or build from source per the building guide. It is a Tauri app on whisper.cpp with GPU acceleration on NVIDIA, AMD and Intel through Vulkan and CoreML. Nobody sells a hosted version.
Who it fits
A person who wants one offline app for files, calls and links, with documents out the other end. A team that needs sharing and search across meetings should look at OpenWhispr.
Limits the README states
Stable timestamps mode is slower. Nothing else is listed.
Sources
- Vibe README and MIT LICENSE: github.com/thewh1teagle/vibe, read 2026-09-16.
More meeting transcription
A meeting assistant under MIT that records your microphone and system audio together, transcribes live on your machine with Whisper or Parakeet, and summarises with a language model you choose.
A desktop app under MIT that transcribes and translates audio and video files, YouTube links and a live microphone offline with Whisper, with speaker identification and SRT, VTT and TXT export.
Added September 16, 2026. Found an error on this page? Write to reveneau@licheo.com and it gets fixed in the next weekly pass.
Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.