Transcribe video to text with word timestamps — TikTok, YouTube, Instagram, X, Twitch. Uses native platform captions when available (~2s) and falls back to Whisper-large-v3-turbo on extracted audio (~30s). When the audio is a known song, returns LRCLIB lyrics + synced timestamps instead of garbled noise.
https://api.hookfindr.com/api/x402/info