
from podcast-transcript-txt-skill31
Convert podcasts from YouTube, Apple Podcasts, and other web sources into cleaned TXT transcripts.
This skill implements a deterministic workflow to find, extract, and clean transcripts from a wide variety of podcast sources. It prioritizes official transcripts, then subtitles, and finally falls back to local ASR (Automatic Speech Recognition) using faster-whisper to ensure a transcript is produced whenever possible.
Use this skill when a user provides a podcast URL (YouTube, Xiaoyuzhou, X/Twitter), a plain episode title, or an audio file and requests a written transcript or text version for reading or analysis.
scripts/podcast_transcript_txt.py handles the end-to-end resolution and extraction pipeline.Compatible with agents that can execute Python scripts and have access to yt-dlp and faster-whisper dependencies.
This skill has not been reviewed by our automated audit pipeline yet.