Align Transcript to Baselines

🤖 AI Drafted (Not reviewed)

Joins a page’s known transcript to its Kraken line baselines, on-device and free — no model runs. Each transcript line is hung on its baseline in reading order, so the reader can anchor the text to the page image. A page whose line count does not match its baseline count is left unaligned rather than have the wrong text placed on it. Run after Detect Regions (Kraken) on pages that already have a transcript.

Tool id	align_transcript
Category	vision
Uses a language model	no
Needs a generative model	no
Runs over many items	no
Item handling	batch
Structured output	no
Human-verified	not yet

What it reads

This tool takes no inputs.

What it emits

Port	Type	Required	What it is
Documents (documents)	json	—	The documents processed, so a downstream step links back.
Aligned (aligned_count)	number	—	How many pages were aligned (exact line-count match).
Skipped (skipped_count)	number	—	Pages left unaligned (no transcript, no baselines, or mismatch).

Options

This tool has no options.

The prompt it sends

This tool does not send a prompt — it does its work in code.