Title:Language Model Supervision for Handwriting Recognition Model Adaptation

Abstract:Training state-of-the-art offline handwriting recognition (HWR) models requires large labeled datasets, but unfortunately such datasets are not available in all languages and domains due to the high cost of manual this http URL address this problem by showing how high resource languages can be leveraged to help train models for low resource this http URL propose a transfer learning methodology where we adapt HWR models trained on a source language to a target language that uses the same writing this http URL methodology only requires labeled data in the source language, unlabeled data in the target language, and a language model of the target language. The language model is used in a bootstrapping fashion to refine predictions in the target language for use as ground truth in training the this http URL this approach we demonstrate improved transferability among French, English, and Spanish languages using both historical and modern handwriting datasets. In the best case, transferring with the proposed methodology results in character error rates nearly as good as full supervised training.