Handwritten Stenography Recognition and the LION Dataset

TL;DR

This paper establishes a baseline for handwritten stenography recognition using the novel LION dataset, demonstrating that integrating stenographic theory and synthetic pre-training significantly improves recognition accuracy.

Contribution

It introduces the LION dataset, applies modern text recognition models to stenography, and explores encoding methods that incorporate stenographic knowledge, advancing the field.

Findings

Baseline CER of 29.81% and WER of 55.14%.

Encoding stenographic features reduces CER to 24.5-26%.

Pre-training with synthetic data further improves accuracy.

Abstract

Purpose: In this paper, we establish a baseline for handwritten stenography recognition, using the novel LION dataset, and investigate the impact of including selected aspects of stenographic theory into the recognition process. We make the LION dataset publicly available with the aim of encouraging future research in handwritten stenography recognition. Methods: A state-of-the-art text recognition model is trained to establish a baseline. Stenographic domain knowledge is integrated by applying four different encoding methods that transform the target sequence into representations, which approximate selected aspects of the writing system. Results are further improved by integrating a pre-training scheme, based on synthetic data. Results: The baseline model achieves an average test character error rate (CER) of 29.81% and a word error rate (WER) of 55.14%. Test error rates are reduced significantly by combining stenography-specific target sequence encodings with pre-training and fine-tuning, yielding CERs in the range of 24.5% - 26% and WERs of 44.8% - 48.2%. Conclusion: The obtained results demonstrate the challenging nature of stenography recognition. Integrating stenography-specific knowledge, in conjunction with pre-training and fine-tuning on synthetic data, yields considerable improvements. Together with our precursor study on the subject, this is the first work to apply modern handwritten text recognition to stenography. The dataset and our code are publicly available via Zenodo.