Is there an easy way to transfer a model's predictions back into fine-tuning html?

[megamattc]

Hello,

I was wondering if there was a convenient way to take the text obtained from applying a model to an image and insert the lines of that text into the correction boxes of the HTML file obtained from the ketos transcribe command? That would substantially accelerate the creation of gold-standard training data for the model.

[mittagessen]

The whole ketos transcribe/extract workflow is deprecated as is the bounding box segmenter. It isn't integrated with the standardized serialization formats.

With the baseline workflow you can just produce ALTO/PageXML and feed them back into any kind of annotator (or just use eScriptorium for the whole workflow in a nice interface).