• Home
  • Search
  • How well does multiple OCR error correction generalize?
  • Open Access IconOpen Access
  • Cite Icon14
  • https://doi.org/10.1117/12.2042502Copy DOI Icon

How well does multiple OCR error correction generalize?

  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

As the digitization of historical documents, such as newspapers, becomes more common, the need of the archive patron for accurate digital text from those documents increases. Building on our earlier work, the contributions of this paper are: 1. in demonstrating the applicability of novel methods for correcting optical character recognition (OCR) on disparate data sets, including a new synthetic training set, 2. enhancing the correction algorithm with novel features, and 3. assessing the data requirements of the correction learning method. First, we correct errors using conditional random fields (CRF) trained on synthetic training data sets in order to demonstrate the applicability of the methodology to unrelated test sets. Second, we show the strength of lexical features from the training sets on two unrelated test sets, yielding a relative reduction in word error rate on the test sets of 6.52%. New features capture the recurrence of hypothesis tokens and yield an additional relative reduction in WER of 2.30%. Further, we show that only 2.0% of the full training corpus of over 500,000 feature cases is needed to achieve correction results comparable to those using the entire training corpus, effectively reducing both the complexity of the training process and the learned correction model.

Similar Papers
  • Research Article
  • Citations27

Adaptive Language Modeling With Varied Sources to Cover New Vocabulary Items

  • May 01, 2004
  • IEEE Transactions on Speech and Audio Processing
  • S.E Schwarm +2
  • Research Article
  • Citations17

Theoretical Analysis of Diversity in an Ensemble of Automatic Speech Recognition Systems

  • Mar 01, 2014
  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • Kartik Audhkhasi +3
  • Research Article
  • Citations22

Joint Discriminative Decoding of Words and Semantic Tags for Spoken Language Understanding

  • Jan 01, 2013
  • IEEE Transactions on Audio, Speech, and Language Processing
  • Anoop Deoras +3
  • Conference Article
  • Citations7

HMM-based sliding video text recognition for Turkish broadcast news

  • Sep 01, 2009
  • Temucin Som +2
  • Conference Article
  • Citations171

Discriminative language modeling with conditional random fields and the perceptron algorithm

  • Jan 01, 2004
  • Brian Roark +3
  • Research Article
  • Citations2

On Reducing Harmonic and Sampling Distortion in Vocal Tract Length Normalization

  • Jan 01, 2013
  • IEEE Transactions on Audio, Speech, and Language Processing
  • Néstor Becerra Yoma +4
  • Conference Article
  • Citations11

Portability of syntactic structure for language modeling

  • May 01, 2001
  • C Chelba
  • Conference Article

Improved Covariance Modeling For Maximum Likelihood Multiple Subspace Transformations

  • Mar 18, 2005
  • Xi Zhou +3
  • Conference Article
  • Citations136

Towards Fast and Accurate Streaming End-To-End ASR

  • May 01, 2020
  • Bo Li +6
  • PDF
  • Research Article
  • Citations10

Auxiliary Networks for Joint Speaker Adaptation and Speaker Change Detection

  • Nov 25, 2020
  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
  • Leda Sari +2
  • Conference Article
  • Citations3

Improving Entity Recall in Automatic Speech Recognition with Neural Embeddings

  • Jun 06, 2021
  • Christopher Li +5
  • Conference Article
  • Citations43

Improvements in hidden Markov model based Arabic OCR

  • Dec 01, 2008
  • Rohit Prasad +4
  • Dissertation

Applying unsupervised grammar induction to OCR error correction

  • Jan 01, 2014
  • Samuel R Scarano
  • Conference Article
  • Citations33

Morphological degradation models and their use in document image restoration

  • Nov 13, 2002
  • Qigong Zheng +1
  • PDF
  • Research Article
  • Citations3

Manual sorting of numerals in an inflective language for language modelling

  • Mar 17, 2014
  • International Journal of Speech Technology
  • Gregor Donaj +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.