ojbduce Transcribe Audio Skill
Purpose
Transcribe audio to text using the LemonFox speech-to-text API. Handles multiple input types so it can be used standalone or as a fallback from other skills.
How to Use
# From a local audio file
python ~/.cursor/skills/transcribe-audio/scripts/transcribe_audio.py --file path/to/audio.mp3
# From a direct audio URL
python ~/.cursor/skills/transcribe-audio/scripts/transcribe_audio.py --audio-url https://example.com/audio.mp3
# From a YouTube video (downloads audio, then transcribes)
python ~/.cursor/skills/transcribe-audio/scripts/transcribe_audio.py --youtube-url https://www.youtube.com/watch?v=VIDEO_ID
Arguments
| Arg | Required | Description |
|---|---|---|
--file |
One of three | Path to a local audio file |
--audio-url |
One of three | Direct URL to an audio file |
--youtube-url |
One of three | YouTube video URL (audio will be downloaded) |
--output |
No | Write transcript to file; if omitted, print to stdout |
--output-format |
No | text (default) or json |
--language |
No | Language code (default: en) |
--api-key-env |
No | Env var name for LemonFox key (default: LEMONFOX_API_KEY) |
Environment
LEMONFOX_API_KEY: Required. Set in the environment; never pass the key as an argument or in prompts.
Exit Codes
0Success1Invalid args2Audio download failure4Transcription API error5Unexpected error
Dependencies
requests(pip install requests)yt-dlp(pip install yt-dlp) — only needed for--youtube-url
For LemonFox API details, see reference.md.