Privacy-first
audio transcription
Powered by OpenAI's Whisper. Runs entirely on your machine. Your audio never leaves your device.
Open Source · MIT License · No cloud · No subscriptions
Powered by OpenAI's Whisper. Runs entirely on your machine. Your audio never leaves your device.
Open Source · MIT License · No cloud · No subscriptions
No compromises. Full control.
MP3, WAV, M4A, OGG, FLAC, AAC, MP4, MOV, MKV and more.
Know who said what. Powered by pyannote with local HF models.
Summaries, action items, translations, custom prompts — all via local LLM.
Paste a YouTube, TikTok, Instagram, Facebook, X, Vimeo, or Twitch link — done.
Drop a folder of files and let it run. Multi-select, bulk export, bulk actions.
JSON, SRT, TXT, and Markdown with speaker runs, metadata, and table of contents.
Record directly from your microphone with pause/resume and auto-transcription.
Search across all transcripts instantly with SQLite FTS5.
Standalone desktop app or self-hosted via Docker — no Python required for the app.
Based on official READMEs as of April 2026.
← scroll to compare →
| Feature | AmicoScript | Buzz | WhisperX | Whisper WebUI | IFW |
|---|---|---|---|---|---|
| Local-only (no cloud) | ✅ | ✅ | ✅ | ✅ | ✅ |
| Speaker diarization | ✅ | ✅ | ✅ | ✅ | ✅ |
| Ollama / LLM integration | ✅ | ❌ | ❌ | ❌ | ❌ |
| URL import (7 platforms) | ✅ | YouTube only | ❌ | YouTube only | ❌ |
| Batch processing | ✅ | ✅ | ❌ | ✅ | ✅ |
| Desktop app (no Python needed) | ✅ | ✅ | ❌ | ❌ | ❌ |
| Docker support | ✅ | ❌ | ❌ | ✅ | ❌ |
| Web UI | ✅ | ❌ | ❌ | ✅ | ❌ |
Docker is the easiest path. Or download the desktop app.
docker compose up --buildOpen http://localhost:8002
pip install -r backend/requirements.txt
python run.pyRequires Python 3.10+
Or skip Python entirely — download the desktop app for Mac or Windows.
Free, open source, and runs entirely on your machine.
Questions? Join the community on Telegram @amicoscript