Buzz
Offline audio and video transcription and translation app for your own computer, built on Whisper.
- Stars
- 22.4k
- Forks
- 1.7k
- Licence
- MIT
- Language
- Python
- Last update
- 4 days ago
- Type
- App
What it is
Buzz turns spoken audio and video into text, and can translate it, without sending your files to a cloud service, according to its description. It is powered by OpenAI's Whisper speech recognition model. People use it to make transcripts and subtitles from recordings, interviews, lectures, or YouTube links.
- Transcribe an audio or video file, or a YouTube link, into text
- Export a transcript to TXT, SRT, or VTT subtitle format
- Transcribe live speech from a microphone and show it in a presentation window
- Use the command-line interface to batch or script transcriptions
- Set a watch folder so new files are transcribed automatically
Great for
- People who want to transcribe recordings without uploading them to an online service
- Creators who need SRT or VTT subtitle files or TXT transcripts
- Users who want live transcription from a microphone during events or presentations
- Users with an Nvidia GPU or Apple Silicon Mac who want faster processing
Not ideal for
- Users on Intel Macs, since the README says the last supported version for them is 1.4.5
- People who need a hosted, no-install web service, because Buzz is installed locally
- Users who expect a fully signed Windows app, since the README says the Windows app is not signed and shows a warning
How to get it
- 1
Download the app
You'll need: ffmpeg (required only for the PyPI install route) · Python 3.12 (required only for the PyPI install route)
Install via Flatpak (Linux)
flatpak install flathub io.github.chidiwilliams.BuzzInstall via PyPI (after installing ffmpeg and Python 3.12)
pip install buzz-captionsRun after PyPI install
python -m buzz - 2
Run it
- Download the installer for your platform from SourceForge, or install the Flatpak or Snap on Linux
- On Windows, select More info and then Run anyway if the unsigned-app warning appears
- Open Buzz and import an audio or video file, or paste a YouTube link
- Choose a transcript export format such as TXT, SRT, or VTT
- 3
Get the most out of it
- Install CUDA acceleration from Help -> About Buzz if you have an Nvidia GPU and are on version 1.4.6 or later, per the README
- Use the live microphone transcription with the presentation window for events
- Set up a watch folder to transcribe new files automatically
- Use the command-line interface for scripting and automation
- Explore plugins such as AI summary generation and automated transcript resizing
Downloads
Strengths
- Runs offline on your own computer, according to the README
- Supports several Whisper backends, including CUDA for Nvidia GPUs, Apple Silicon, and Vulkan
- Offers multiple export formats and a transcript viewer with search and playback speed control
Trade-offs
- Some acceleration options depend on specific hardware and versions; the README says CUDA setup applies to version 1.4.6 or later, which it describes as not yet released at the time of the README
- The Windows installer is unsigned, so installation shows a security warning
- The README is truncated, so detailed setup and model requirements are unclear from this text
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| buzz (this) | 22.4k | App | beginner | 4 days ago | People who want to transcribe recordings without uploading them to an online service |
| whisper.cpp | 54.9k | Library | intermediate | 7 days ago | Developers who want to embed offline speech-to-text in a C, C++, or other native app |
| faster-whisper | 26.4k | Library | intermediate | 5 days ago | Python developers who need to add speech-to-text to an application |
| subtitleedit | 14.8k | App | beginner | 4 days ago | People who need to fix timing or text in existing subtitle files |
| voicebox | 59.2k | App | beginner | 30 days ago | People who want text-to-speech or voice cloning without sending audio to cloud services |
| meetily | 32.4k | App | beginner | 4 days ago | People who want meeting notes without sending recordings to a cloud service |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.