yap

Fast on-device speech-to-text transcription on macOS 26+ using Apple Speech.framework, supporting multiple languages and output formats without model downloads.

Use yap for fast on-device transcription on macOS using Apple's Speech.framework.

Quick start

yap transcribe /path/to/audio.mp3
yap transcribe /path/to/audio.m4a --locale de-DE
yap transcribe /path/to/video.mp4 --srt -o captions.srt

Options

• --locale <locale> — Language locale (e.g., de-DE, en-US, zh-CN)
• --censor — Redact certain words/phrases
• --txt / --srt — Output format (default: txt)
• -o, --output-file — Save to file instead of stdout

Advantages over Whisper

• Native Apple Speech.framework (optimized for Apple Silicon)
• No model download required
• Faster processing
• Lower memory usage

Notes

• Requires macOS 26 (Tahoe) or later
• Supported languages depend on installed Apple Speech models