vlayervlayer docs
Data lookups

Transcripts

Read the spoken text of an Instagram post or reel, a TikTok or YouTube video, or an X post.

Transcript

  • Instagram. Input: post or reel URL. Returns one entry per carousel item: id, shortcode and text, which is null when there is no speech. Transcripts are generated by AI and take 10 to 30 seconds, so a call can time out; retry it. They cover videos under 2 minutes.
  • TikTok. Input: video URL and a two-letter language code. Returns the id, URL and the transcript as WebVTT, with timestamps.
  • YouTube. Input: one video. Returns the video's transcript. It reads one video at a time.
  • X. Input: one post. Returns the post's transcript.

Last updated on

On this page