Documentation / Auto Captions

Auto Captions

Auto Captions

Auto captions are the only AI-powered part of EzClap: the audio from your clip is transcribed with speech-to-text, you review and edit the result in the browser, and the captions are burned into the final video in your chosen caption style.

Generating captions

  1. From the Clips dashboard, find a processed clip and click generate captions on its card. (You can also start generation from inside the caption editor if a clip has none yet.)
  2. EzClap transcribes the clip's audio. The transcription language is set in Settings → captions — auto-detect by default, or force a specific language such as English, French, German, Spanish, Italian, Portuguese, or Dutch.
  3. When the job finishes, the clip card switches to download with captions, and you can open the caption editor to review the text.

Free accounts run caption jobs manually, one at a time. Pro / Partner accounts can enable Auto-Generate Captions in Settings so every clip is captioned immediately after processing, and get access to Google Fonts and higher quality output.

Editing captions

The caption editor shows a video preview on the left and your caption blocks on the right. Each block is one on-screen caption with its own timing and text.

Per block you can:

  • Edit the text directly in the block's text area.
  • Retime it — play the video to the right moment and use the Start → and ← End buttons to snap the block's start or end to the current frame. Manually timed blocks show a lock icon so automatic re-timing won't touch them.
  • Mark a second speaker — toggle the v1 button to v2 to render that block in your style's voice 2 color.
  • Ghost it — hide the text during render without deleting the timing.
  • Delete it, or insert a new block above or below.
  • Attach a sound to a specific word for word-synced sound effects.

The editor also includes trim controls (set trim start/end, apply the trim, or duplicate the trim as a new clip), a remove-facecam option, and a title field at the top used for your YouTube / TikTok upload title.

When you're done, use apply changes & preview to see the result, then save & close. The regenerate captions button re-runs transcription and overwrites your current captions, and you can download the raw VTT file at any time.

Burning captions in

Click download with captions on the clip card to render the captions permanently into the video using the active caption style. Once a burnt-captions video exists, the editor shows it side by side with the original for comparison, with its own download and delete buttons. If you change captions afterward, re-burn the video to update it.

Which style is used follows the assignment hierarchy — game assignment first, then clip template assignment, then your default style. Set this up in caption styles.

related