Medieval Anglicana Legal Texts for Handwritten Text Recognition, 13th-14th Century

FLOW Project at Bielefeld University created this dataset of 21,579 lines of high-quality Ground Truth data for Handwritten Text Recognition. The data consists of image snippets paired with transcriptions from medieval administrative and legal documents. The dataset was last updated on 2026-04-24.

Use Cases

• Train Handwritten Text Recognition models based on the high-quality line-level image-text pairs.
• Benchmark OCR/HTR performance on medieval scripts based on the provided ground truth transcriptions.
• Study the paleography and language of 13th-14th century English legal documents based on the transcribed content.
• Develop multimodal models for historical document understanding based on the paired image and text data.

Strengths

• Contains 21,579 lines of high-quality ground truth data, providing a substantial training resource.
• Created by a named academic project (FLOW Project at Bielefeld University) for a specific research purpose.
• Focuses on a well-defined historical period (13th-14th century) and document type (Anglicana legal texts).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known, but file formats, sample data, and license information are unknown.
• Data may reflect temporal and linguistic bias inherent to its specific historical source material.

Provenance

Source

FLOW Project at Bielefeld University (Digital History).

Collection Method

Created to facilitate training of HTR models, likely involving manual transcription and image segmentation.

Time Range

13th-14th century.

Freshness

Last updated 2026-04-24 10:00:46; freshness should be verified.

Geography

Likely England, given the 'Anglicana' script and legal context, but not explicitly stated.