Get transcripts from YouTube, TikTok, Instagram Reels, X/Twitter videos and podcasts in one run: plain text, timestamped segments, SRT and VTT. Native captions first, AI speech-to-text fallback, optional AI summary, key points and chapters. Failed URLs are free.
First release: YouTube (videos, Shorts, playlists, channels), TikTok, Instagram Reels, X/Twitter videos, podcast RSS feeds, Apple Podcasts links and direct audio/video files.
Native captions first (YouTube manual, auto and translated captions, platform subtitles, Podcasting 2.0 transcripts), with AI speech-to-text fallback through Apify's OpenRouter proxy (Gemini 2.5 Flash Lite by default) or Whisper with your own OpenRouter key.
Output as text, timestamped segments, SRT and VTT, with optional saved subtitle files.
Optional AI summary, key points, chapters with timestamps, quotes and topics in any language.
Per-URL error rows with codes and hints (free), automatic residential proxy fallback, model fallback on rate limits, and migration-safe resume.
Fixes before publishing: a missing fallback speech-to-text model no longer disables speech-to-text for the whole run; sttModel now reports the model that actually transcribed each item; clearer access-refused messages; long-audio speech-to-text marked Beta.
