It would be helpful if you had a feature where clicking on a word would highlight where you were in the transcription and vice versa. I feel like it would be especially useful with big documents and difficult-to-read handwriting.
1 Like
Absolutely. We’re working on this! 
1 Like
I’m curious about this! Forgive my ignorance, but would this require the transcript to be mapped to coordinates within the image or PDF? That could be a boon to users who are looking to embed the transcript as a text layer.
Yes! We’re actively working on a “spotlighting” feature for the next version of our model. It will automatically detect the coordinates in the image associated with a given line of text in the transcription, so that you’ll be able to hover over one to highlight the associated section of the other. This data will be exportable via XML so you’ll be able to use those coordinates downstream e.g. for embedding the text in the images.
1 Like