OCR Post Correction for Endangered Language Texts
This addresses the challenge of making textual data in endangered languages machine-readable for NLP applications, though it is incremental as it builds on existing OCR methods.
The paper tackled the problem of extracting text from scanned books in critically endangered languages where general-purpose OCR tools perform poorly due to data scarcity, and developed a tailored OCR post-correction method that reduced the recognition error rate by 34% on average across three languages.
There is little to no data available to build natural language processing models for most endangered languages. However, textual data in these languages often exists in formats that are not machine-readable, such as paper books and scanned images. In this work, we address the task of extracting text from these resources. We create a benchmark dataset of transcriptions for scanned books in three critically endangered languages and present a systematic analysis of how general-purpose OCR tools are not robust to the data-scarce setting of endangered languages. We develop an OCR post-correction method tailored to ease training in this data-scarce setting, reducing the recognition error rate by 34% on average across the three languages.