Buzz
Offline transcription and translation software runs Whisper on Windows, Linux and macOS, with microphone capture and subtitle export.
Buzz transcribes and translates speech on your own computer using OpenAI's Whisper. It's for people who need transcripts or subtitles from recordings, plus live captions from a microphone. Local transcription works offline; the optional OpenAI Whisper API sends audio to a cloud service.
You can import audio and video files or transcribe YouTube links. Buzz supports transcription in the spoken language and translation into English across many languages. It exports CSV, SRT, TXT and VTT files, so transcripts can serve as readable documents or timed captions.
Buzz Classic is open source under the MIT license, and available for Windows, Linux and macOS. It supports Whisper, Whisper.cpp, Faster Whisper and Whisper-compatible Hugging Face models, as well as the OpenAI Whisper API. GPU acceleration includes NVIDIA CUDA and Vulkan through Whisper.cpp, including support for integrated GPUs.
For difficult recordings, Buzz can separate speech before transcription and identify speakers. Its transcript viewer includes search, audio playback and playback speed controls. A presentation window displays live captions during events, while a watched folder can automatically transcribe new files. Live transcription speed depends on your hardware, language and model size.
The separate macOS app has a native interface, supports Whisper.cpp models and the OpenAI Whisper API, and includes transcript search, audio playback and inline editing.
