Transcript
Download or transcribe text from any audio/video source.
by notegpt-official · MIT
The skill
Download or transcribe text from any audio/video source. For YouTube URLs, fetches subtitles (manual preferred, then auto-generated, then Whisper fallback). For any audio/video file, transcribes directly with faster-whisper — generating plain text, SRT/VTT subtitles, JSON, or timestamped segments. Supports batch processing and translation. Trigger keywords: transcribe, transcript, download transcript, get captions, get subtitles, extract text from video, convert audio to text, YouTube transcript, generate subtitles, transcribe podcast, audio to text, video to text.
Transcript — Universal Audio/Video to Text
Fetch YouTube transcripts via yt-dlp, or transcribe any audio/video file with OpenAI Whisper.
Routing
User input:
├── YouTube URL? → Read references/youtube-transcript.md and follow its workflow
└── Audio/video file? → Read references/whisper-transcription.md and follow its workflow
Always ask the user before large downloads (audio files, Whisper models). Clean up temporary files when done.
Use this skill
Connect your agent to the Skill Store MCP server and ask for it by ref: notegpt-official/transcript-57434982126.
Details
- Publisher
- notegpt-official
- Category
- Creative & Media › Audio & music
- License
- MIT
- Updated
- 2026-10-12
References
More from notegpt-official
- Ai Documents Create, edit, or improve documents according to the user's requirements, always generating a file.
- Ai Sheets Create, improve, or analyze spreadsheets according to the user's requirements, always generating…
- Browser Automation Automate browser tasks that require bypassing bot detection — scrape protected pages, fill forms…
- Data Analysis Analyze the user's data and generate easy-to-understand visual insight reports with professional…