Text Recognition Tools

The National Archives has developed text recognition tools that can convert old handwriting into machine-typed text. The content recognition tool works with handwriting from the 1600s to the mid-1900s. There are separate models for reading Latin and Cyrillic script.

Handwriting recognition is increasingly being incorporated into the National Archives' services, and the demo currently available is still in its pilot stage.

How to proceed:

1. Download an image of the document to your computer from Astia.
2. Go to the text recognition tool's website. Choose whether you want to use the model that reads Latin script or the model that reads Cyrillic script.
3. Drag the image with your mouse into the field on the left side of the tool's window (you can also click the field to open the file via File Explorer).
4. Click "Process image." The tool will recognize the text content of the image and convert it into readable form.

Go to the Latin-script text recognition model here.

Go to the Cyrillic-script text recognition model here.