Skip to content
Home›AI & agents›AI tools›Buzz transcription
Transcribe & subtitle

Buzz: a free desktop app for offline transcripts and subtitles

A free, MIT-licensed desktop app for offline transcription and translation that exports TXT, SRT and VTT [2][1].

Does
Transcripts and subtitles from a desktop app, no code
Best for
Non-technical vloggers who want local captions
Cost
Free (MIT licence) [1]
Runs
Desktop app, locally
Open the repositoryOfficial siteQuick start
Updated 2 Nov 2026

Import a file or YouTube link, pick a model, export [2]. It also offers live microphone transcription, speaker identification, a watch folder and a command line [2].

What it does

Licence
MIT [1]
Export formats
TXT, SRT, VTT [2]
Input
Audio, video or YouTube links [2]
Platforms
macOS .dmg (Apple silicon), Windows installer, Linux Flatpak, Snap or AppImage, PyPI [2]
Engines
Whisper, whisper.cpp (Vulkan GPU support), faster-whisper, Whisper-compatible Hugging Face models, OpenAI Whisper API [3]
Offline use
Works without internet once models are downloaded or copied across [4]
Homebrew
buzz cask, version 1.4.5 [5]

Integrations

InterfaceStatusDetail
MCP serverMCP · None foundNo built-in MCP verified
Command lineCLI · Built inDesktop app / scripts
API / libraryAPI · None foundNone documented. Buzz can send audio to the OpenAI API with your key [2][3].

Requirements and setup

  • A CPU with AVX2 [4].
  • macOS: Apple silicon; 1.4.5 was the last Intel release [2].
  • PyPI install: ffmpeg and Python 3.12 [2].
  • faster-whisper engine: NVIDIA GPU with 6GB+ VRAM suggested [4].

Licence, cost and privacy

  • Local engines run on your computer; the OpenAI API option sends audio to a remote server [4].
  • Model cache: ~/Library/Caches/Buzz (macOS), ~/.cache/Buzz (Linux), %USERPROFILE%\AppData\Local\Buzz\Buzz\Cache (Windows) [4].

Limitations

  • Unsigned Windows installer: choose More info, then Run anyway [2].
  • Live transcription is resource-heavy and may lag behind real time [3].
  • Large-V3 can invent words that are not in the audio [4].
  • Models with .En in the name are English only [4].
  • A corrupt model download can crash Buzz; delete it and download again [4].
Quick start

Quick start

On Windows, use the SourceForge installer instead [2].

Copied from the documentation · macOS · source [5]
brew install --cask buzz
Copied from the documentation · Linux (Flatpak) · source [2]
flatpak install flathub io.github.chidiwilliams.Buzz
Copied from the documentation · Linux (Snap) · source [2]
sudo apt-get install libportaudio2 libcanberra-gtk-module libcanberra-gtk3-module
sudo snap install buzz
Copied from the documentation · Python 3.12 · source [2]
pip install buzz-captions
python -m buzz

Open a file, choose a model, export SRT or VTT [2]. On Macs and PCs without an NVIDIA GPU, the FAQ suggests whisper.cpp [4].

Uses for vloggers

  • Vlog captions without the terminal.
  • Searchable interview transcripts.
  • Live on-screen captions at events, via the presentation window [2].
  • A watch folder that transcribes new files as they arrive [2].

Workflows that use it

Some setupLocal AI subtitles

Turn vlog speech into SRT and VTT subtitle files on your own computer, with an app or the command line, then check every name.

Buzz transcriptionAuto Subs (tmoroney)whisper.cppFFmpeg / ffprobe

Alternatives

Transcribe & subtitlewhisper.cpp

Run efficient Whisper inference on local hardware

Free codeOn your computerDeveloperMIT licence
MCP · CommunityCLI · Built inAPI · Built in
Transcribe & subtitleAuto Subs (tmoroney)

Generate speech subtitles inside Resolve, Premiere or After Effects workflows

Free codeOn your computerSome setupMIT licence
MCP · None foundCLI · Built inAPI · None found
Transcribe & subtitleWhisper (OpenAI)

Speech recognition and subtitles for local post-production

Free codeOn your computerDeveloperMIT licence
MCP · CommunityCLI · Built inAPI · Built in
Transcribe & subtitlefaster-whisper

Faster Whisper-based transcription with CTranslate2

Free codeOn your computerDeveloperMIT licence
MCP · CommunityCLI · CommunityAPI · Built in

Sources

  1. raw.githubusercontent.com/chidiwilliams/buzz/main/LICENSE fetched 2 Nov 2026
  2. raw.githubusercontent.com/chidiwilliams/buzz/main/README.md fetched 2 Nov 2026
  3. Introduction | Buzz fetched 2 Nov 2026
  4. FAQ | Buzz fetched 2 Nov 2026
  5. Homebrew Formulae: buzz fetched 2 Nov 2026