The OpenAI Video Transcription skill provides Claude with a robust mechanism for extracting and transcribing audio from a wide range of sources, including YouTube, Bilibili, Douyin, and local media files. By combining yt-dlp for reliable media extraction and OpenAI's Whisper model for high-accuracy speech-to-text, it serves as a high-performance fallback or primary tool for generating subtitles and transcripts. This skill is essential for developers and researchers who need to process multimedia content for analysis, accessibility, or documentation purposes within their AI-driven workflows.
주요 기능
01Processes local audio and video files including MP3, MP4, MOV, and WAV
0228 GitHub stars
03Supports URLs from YouTube, Bilibili, Douyin, and Xiaoyuzhou podcasts
04Returns detailed metadata including language detection and duration
05Generates both raw text transcripts and standardized SRT subtitle files
06Includes intelligent fallback logic for high-reliability transcription requests
사용 사례
01Transcribing research interviews or podcasts directly from URLs into text
02Automating the creation of subtitles for video content across multiple platforms
03Extracting textual data from local media files for archival and searchability