Buzz
brew install --cask buzz
v1.4.4
Transcribe and translate audio files offline using OpenAI's Whisper model.
Why you might care
Buzz is open-source and runs entirely on your machine without sending audio to external servers, making it private and fast. It supports GPU acceleration (including Apple Silicon) and can transcribe from files, YouTube links, or live microphone input with speaker identification.
154
30-day installs · #1100
240
90-day · #1574
1.6k
365-day · #1128
19.8k
★ GitHub stars · updated today
GitHub topics
whisper
Links
- https://github.com/chidiwilliams/buzz
- GitHub: chidiwilliams/buzz
- Brew formula source: Casks/b/buzz.rb
Blurb generated by claude-haiku-4-5 on today.
Raw metadata
{
"alternatives": [
"Whisper Transcription",
"MacWhisper",
"Otter.ai"
],
"artifacts": [
{
"app": [
"Buzz.app"
],
"target": "/Applications/Buzz.app"
},
{
"zap": [
{
"trash": [
"~/Library/Caches/Buzz",
"~/Library/Logs/Buzz",
"~/Library/Preferences/com.chidiwilliams.buzz.plist",
"~/Library/Saved Application State/com.chidiwilliams.buzz.savedState"
]
}
]
}
],
"auto_updates": 1,
"categories": [
"dev-tools",
"audio",
"productivity"
],
"deprecated": 0,
"deprecation_reason": null,
"desc": "Transcribe and translate audio",
"disable_reason": null,
"disabled": 0,
"display_name": "Buzz",
"enrichment_fetched_at": "2026-06-20T22:41:16+00:00",
"first_seen": "2026-06-20T00:47:34+00:00",
"full_token": "buzz",
"github_default_branch": "main",
"github_last_commit_at": "2026-06-20T09:16:14Z",
"github_readme_excerpt": "[[\u7b80\u4f53\u4e2d\u6587](readme/README.zh_CN.md)] \u003c- \u70b9\u51fb\u67e5\u770b\u4e2d\u6587\u9875\u9762\u3002\n\n# Buzz\n\n[Documentation](https://chidiwilliams.github.io/buzz/)\n\nTranscribe and translate audio offline on your personal computer. Powered by\nOpenAI\u0027s [Whisper](https://github.com/openai/whisper).\n\n\n[](https://github.com/chidiwilliams/buzz/actions/workflows/ci.yml)\n[](https://codecov.io/github/chidiwilliams/buzz)\n\n[](https://GitHub.com/chidiwilliams/buzz/releases/)\n\n\n\n## Features\n- Transcribe audio and video files or Youtube links\n- Live realtime audio transcription from microphone\n - Presentation window for easy accessibility during events and presentations\n- Speech separation before transcription for better accuracy on noisy audio\n- Speaker identification in transcribed media\n- Multiple whisper backend support\n - CUDA acceleration support for Nvidia GPUs\n - Apple Silicon support for Macs\n - Vulkan acceleration support for Whisper.cpp on most GPUs, including integrated GPUs\n- Export transcripts to TXT, SRT, and VTT\n- Advanced Transcription Viewer with search, playback controls, and speed adjustment\n- Keyboard shortcuts for efficient navigation\n- Watch folder for automatic transcription of new files\n- Command-Line Interface for scripting and automation\n- Plugin system with plugins like AI summary generation and automated transcript resizing\n\n## Installation\n\n### macOS\n\nDownload the `.dmg` from the [SourceForge](https://sourceforge.net/projects/buzz-captions/files/).\n\n### Windows\n\nGet the instal",
"github_repo": "chidiwilliams/buzz",
"github_stars": 19775,
"github_topics": [
"whisper"
],
"homepage": "https://github.com/chidiwilliams/buzz",
"homepage_og_description": null,
"homepage_og_image": null,
"homepage_title": null,
"installs_30d": 154,
"installs_365d": 1571,
"installs_90d": 240,
"last_seen": "2026-06-20T00:47:34+00:00",
"llm_generated_at": "2026-06-20T23:06:50+00:00",
"llm_model": "claude-haiku-4-5",
"names": [
"Buzz"
],
"one_liner": "Transcribe and translate audio files offline using OpenAI\u0027s Whisper model.",
"rank_30d": 1100,
"rank_365d": 1128,
"rank_90d": 1574,
"raw_hash": "693f05a0746824d2",
"ruby_source_path": "Casks/b/buzz.rb",
"tap": "homebrew/cask",
"token": "buzz",
"version": "1.4.4",
"why_use_this": "Buzz is open-source and runs entirely on your machine without sending audio to external servers, making it private and fast. It supports GPU acceleration (including Apple Silicon) and can transcribe from files, YouTube links, or live microphone input with speaker identification."
}