We’ve given the Transkribus How to Guides a reboot! New and improved guidelines are here to help Transkribus users achieve the automated transcription and full-text search of their collections. These guidelines reflect the functionalities of the latest version of Transkribus, with fresh features like improved line segmentation and powerful searching technology. All the guides can be found on the Transkribus wiki.
If you’re new to Transkribus, and want to try creating training data for Automated Text Recognition, you can start with our Simple How to Guide.
Whilst information laid out in tables and forms may look neat to the human eye, it can constitute problems when it comes to the Automated Text Recognition of historical documents. Our new Tables How to Guide can help!
Maybe you already have existing transcripts at your archive or as part of a research project. If so, you can use these transcripts to train a model to recognise the writing in your collection.
Once you have trained a model to recognise the documents in your collection, the Keyword Spotting tool in Transkribus allows you to conduct powerful full-text searches. Find out more about Keyword Spotting in our guide.
Finally, if you are more interested in creating rich transcripts in Transkribus as part of a scholarly edition, our Advanced How to Guide is the one for you.
We are working hard on more user guides to help Transkribus users work with the platform as efficiently as possible. In the meantime, we welcome any feedback by email.