Audio Transcribe
v3Auto-transcribe voice messages locally using faster-whisper with selectable Whisper 模型s, no API key required.
by 运行时依赖
安装命令
点击复制技能文档
Audio Transcription 技能
Auto-transcribe voice messages using faster-whisper (local, no API key needed).
Requirements pip 安装 faster-whisper
模型s 下载 automatically on first use.
Usage Transcribe a file python3 /root/clawd/技能s/audio-transcribe/scripts/transcribe.py /path/to/audio.ogg
Change 模型 (edit script)
Edit transcribe.py and change:
模型 = Whisper模型('small', device='cpu', compute_type='int8') # Options: tiny, base, small, medium, large-v3
模型s 模型 Size VRAM/RAM Speed Use Case tiny 39 MB ~1 GB ⚡⚡⚡ Quick drafts base 74 MB ~1 GB ⚡⚡ Basic accuracy small 244 MB ~2 GB ⚡ Recommended medium 769 MB ~5 GB 🐢 Better accuracy large-v3 1.5 GB ~10 GB 🐢🐢 Best accuracy Integration
Clawd机器人 auto-transcribes incoming voice messages when this 技能 is enabled.
Files scripts/transcribe.py — MAIn transcription script 技能.md — This file