Buzz
github.com
- Free plan
- Windows
- macOS
- Linux
What Buzz is
Buzz is a free desktop application for transcribing and translating audio offline on a personal computer using OpenAI's Whisper. It runs on Linux, macOS and Windows, and accepts audio and video files, YouTube links and live microphone audio. Its tools include speech separation, speaker identification, a searchable transcript viewer, keyboard shortcuts and a watch folder for automatic transcription. Transcripts can be exported as TXT, SRT or VTT.
Buzz supports Whisper, Whisper.cpp, Faster Whisper and Whisper-compatible models from Hugging Face. It can work without an internet connection after the required models are downloaded and copied to the offline computer. Alternatively, setting an API key in Preferences lets it use the OpenAI API for remote speech recognition. A command-line interface supports scripting and automation, and a plugin system has examples such as AI summaries and transcript resizing. Smaller models run faster but may be less accurate; larger ones need more capable hardware or more transcription time. Buzz requires AVX2 CPU support. On macOS, it requires Apple silicon; version 1.4.5 was the last version supporting Intel Macs.
Who it suits
Buzz suits people who want a free desktop tool for local transcription and translation, including those using scripts or plugins. It is not suited to very old computers without AVX2 or Intel Macs needing a newer version.
Good to know, and what to watch
- Works with audio, video, YouTube links and live microphone audio.
- Exports transcripts as TXT, SRT or VTT.
- Can run offline after models are transferred to the computer.
- Includes a command-line interface and plugin system.
- Requires AVX2 CPU support.
- Current macOS support requires Apple silicon.
- Smaller models may be less accurate.
- Unsigned Windows app may trigger an installation warning.
Plans
Every plan and priceBuzz
Free plan
Getting started
- Open the Buzz website on GitHub.
- Choose the version for Linux, macOS, or Windows.
- Download the required models; for offline use, copy them to the computer that will run Buzz.
- For remote speech recognition, set an OpenAI API key in Preferences.
Limits to check
There is no paid plan listed. Smaller models run faster but can be less accurate; larger models need more capable hardware or more transcription time.
The verdict
Buzz offers offline transcription, multiple model choices and export formats at no cost. Check the AVX2 and macOS hardware requirements, and expect a warning during Windows installation.
Questions people ask
How much does Buzz cost?
Buzz is free. Its listed plan is free.
Which operating systems does it support?
Buzz supports Linux, macOS, and Windows. Current macOS support requires Apple silicon; version 1.4.5 was the last version supporting Intel Macs.
Can Buzz work without an internet connection?
Yes. It can run offline after the required models have been downloaded and copied to the offline computer.
What formats can transcripts be exported in?
Buzz exports transcripts as TXT, SRT, or VTT files.
Is Buzz open source?
Buzz is distributed under the MIT License.
What computers can run Buzz?
Buzz requires AVX2 CPU support, so very old computers are not supported.
What the maker publishes
- API option
- Buzz can use the OpenAI API for speech recognition on a remote server when an API key is set in Preferences. source
- Automation
- Buzz includes a command-line interface for scripting and automation, and a plugin system with examples such as AI summaries and transcript resizing. source
- Export formats
- TXT,SRT,VTT source
- Exports
- Transcripts can be exported as TXT, SRT, or VTT files. source
- Free plan
- Yes source
- Input
- It can transcribe audio and video files, YouTube links, and live microphone audio. source
- License
- The repository is distributed under the MIT License. source
- Mac support
- The README says Buzz now requires Apple silicon on macOS; version 1.4.5 was the last version supporting Intel Macs. source
- Models
- Buzz supports Whisper, Whisper.cpp, Faster Whisper, and Whisper-compatible models from Hugging Face. source
- Model tradeoffs
- The FAQ says smaller models run faster but can be less accurate, while larger models require more capable hardware or more transcription time. source
- Offline use
- Buzz can run without an internet connection after the required models are downloaded and copied to the offline computer. source
- Purpose
- Buzz transcribes and translates audio offline on a personal computer using OpenAI's Whisper. source
- Speaker identification
- Yes source
- System requirements
- Buzz requires AVX2 CPU support, so very old computers are not supported. source
- Timestamp support
- Yes source
- Transcription tools
- Features include speech separation, speaker identification, a searchable transcript viewer, keyboard shortcuts, and a watch folder for automatic transcription. source
- Windows note
- The README warns that the Windows app is unsigned and may trigger an installation warning. source
Checked against the maker's pages on 7 Oct 2026.