Filters








3,649 Hits in 6.8 sec

N-gram language models for document image decoding

Gary E. Kopec, Maya R. Said, Kris Popat, Paul B. Kantor, Tapas Kanungo, Jiangying Zhou
<span title="2001-12-18">2001</span> <i title="SPIE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ttq7pbic7jboja6n4lpaxbsfuu" style="color: black;">Document Recognition and Retrieval IX</a> </i> &nbsp;
Probabilistic character n-grams (n=2-5) are used in a two-pass strategy where the decoder first uses a very weak language model to generate a lattice of candidate output strings.  ...  This paper explores the problem of incorporating linguistic constraints into document image decoding, a communication theory approach to document recognition.  ...  The authors wish to thank Dan Bloomberg and Les Niles for many stimulating discussions throughout the course of this work.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.450728">doi:10.1117/12.450728</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/drr/KopecSP02.html">dblp:conf/drr/KopecSP02</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/m66eb44mibfjjc6u3hvzr6hvua">fatcat:m66eb44mibfjjc6u3hvzr6hvua</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170810132932/http://rleweb.mit.edu/dspg/documents/N-GramLanguageModels.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c1/6f/c16f8ce7ee832cd3da204a44769aab4dbbf0f87a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.450728"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Separating Optical and Language Models Through Encoder-Decoder Strategy for Transferable Handwriting Recognition

Adeline Granet, Emmanuel Morin, Harold Mouchere, Solen Quiniou, Christian Viard-Gaudin
<span title="">2018</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/htymowumwvfj5g7qaexx3w2rti" style="color: black;">2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR)</a> </i> &nbsp;
The language decoder creates from a vector of letter-n-grams a word as a sequence of characters.  ...  The optical encoder transforms the input word image into a non-latent space that depends only on the letter-n-grams: it enables it to be independent of the language.  ...  ENCODER-DECODER FOR TRANSFER LEARNING Based on the work on image description generation, we define a system that is divided into two complementary components: an optical model and a language model, as  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/icfhr-2018.2018.00061">doi:10.1109/icfhr-2018.2018.00061</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/icfhr/GranetMMQV18.html">dblp:conf/icfhr/GranetMMQV18</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wrnhsjcj2vc5hetk3b3j7uj3t4">fatcat:wrnhsjcj2vc5hetk3b3j7uj3t4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190502225343/https://hal.archives-ouvertes.fr/hal-01821598/document" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/06/37/06371a61d35bb71e04f7607579562ddfc73494e3.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/icfhr-2018.2018.00061"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Transcription of Spanish Historical Handwritten Documents with Deep Neural Networks

Emilio Granell, Edgard Chammas, Laurence Likforman-Sulem, Carlos-D. Martínez-Hinarejos, Chafic Mokbel, Bogdan-Ionuţ Cîrstea
<span title="2018-01-11">2018</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/au3ye363lzbdroopx7rzfyv63m" style="color: black;">Journal of Imaging</a> </i> &nbsp;
The digitization of historical handwritten document images is important for the preservation of cultural heritage.  ...  It consists of associating a powerful optical recognition system that will cope with image noise and variability, with a language model based on sub-lexical units that will model OOV words.  ...  Figure 9 . 9 Results obtained by decoding at the HMM character level by using n-gram language models with size n = {1, . . . , 15}.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/jimaging4010015">doi:10.3390/jimaging4010015</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5wx5p7nasvg7rjko7dxptnja7i">fatcat:5wx5p7nasvg7rjko7dxptnja7i</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180726222631/https://res.mdpi.com/def5020027b6e808b9357f0d76041251e3a6a896fdaf797c01958cd008c64abe9c915de63e4e3627e039d841c6fe0103c5e228bdc6e0eeaac06a302d332e7848490b6ef97a1968eb58fe149d0a253c9fa73b0875de0884aad4473e7d9487509cb9cbf89e7b0360ec7dda15ed7ae02de8aa0fb9c98603151bc33e9fddd2acdfe7b49d95393585ade9fdb9523256ae2da193a507?filename=&amp;attachment=1" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/9d/31/9d31aea32d1325f10b32feff5c4c0c6fc3ad972b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/jimaging4010015"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

Duplicate detection for symbolically compressed documents

Dar-Shyang Lee, J.J. Hull
<span title="">1999</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/bytixui4szbgzkqzgwf65jphty" style="color: black;">Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR &#39;99 (Cat. No.PR00318)</a> </i> &nbsp;
These techniques are specifically designed for binary document images.  ...  This paper describes a method for duplicate detection on symbolically compressed document images.  ...  Markov models have been used for natural language modeling.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/icdar.1999.791785">doi:10.1109/icdar.1999.791785</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/icdar/LeeH99.html">dblp:conf/icdar/LeeH99</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/adij7apekrhttg7zwyiqsv6mzm">fatcat:adij7apekrhttg7zwyiqsv6mzm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20160606115312/http://jonathanjhull.com/content/pubs/lee_icdar99.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f0/4e/f04e87b08226de302bb14907602abd2bb90d5115.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/icdar.1999.791785"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

An iterative multimodal framework for the transcription of handwritten historical documents

Vicent Alabau, Carlos-D. Martínez-Hinarejos, Verónica Romero, Antonio-L. Lagarda
<span title="">2014</span> <i title="Elsevier BV"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/6r4znskbk5h2ngu345slqsm6eu" style="color: black;">Pattern Recognition Letters</a> </i> &nbsp;
In the two alternatives similar models (Hidden Markov Models and n-grams) and decoding processes (Viterbi decoding) are employed, which allows a possible combination of the two modalities with little difficulties  ...  One alternative for transcribing the ancient manuscripts is the use of speech dictation by using Automatic Speech Recognition techniques.  ...  for speech, strokes for handwritten text) and n-grams language models (LM) are used to model word sequences (Jelinek, 1998) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.patrec.2012.11.007">doi:10.1016/j.patrec.2012.11.007</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/xgujgy2mebefnhunkgzoaekb3y">fatcat:xgujgy2mebefnhunkgzoaekb3y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170922083906/https://riunet.upv.es/bitstream/handle/10251/49463/htr-dictation.pdf?sequence=3" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c6/99/c699fe16f72e803e1509de81dd02a08b307a8030.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.patrec.2012.11.007"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> elsevier.com </button> </a>

Lexically Aware Semi-Supervised Learning for OCR Post-Correction [article]

Shruti Rijhwani, Daisy Rosenblum, Antonios Anastasopoulos, Graham Neubig
<span title="2021-11-04">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In addition, to enforce consistency in the recognized vocabulary, we introduce a lexically-aware decoding method that augments the neural post-correction model with a count-based language model constructed  ...  Much of the existing linguistic data in many languages of the world is locked away in non-digitized books and documents.  ...  Editors for feedback on the paper.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.02622v1">arXiv:2111.02622v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3cs65eyejrh2dd2cri4nnnt45y">fatcat:3cs65eyejrh2dd2cri4nnnt45y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211108232842/https://arxiv.org/pdf/2111.02622v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/87/9d/879d8318be60d36d0132ef1181f265afddcdcf21.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.02622v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Detecting duplicates among symbolically compressed images in a large document database

Dar-Shyang Lee, Jonathan J. Hull
<span title="">2001</span> <i title="Elsevier BV"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/6r4znskbk5h2ngu345slqsm6eu" style="color: black;">Pattern Recognition Letters</a> </i> &nbsp;
An algorithm for duplicate document detection is proposed in this paper that operates directly on images that have been symbolically compressed using techniques related to the ongoing JBIG2 standardization  ...  The detection of duplicate images is a useful means of indexing a large database of documents.  ...  Markov models have been used for natural language modeling.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/s0167-8655(00)00115-x">doi:10.1016/s0167-8655(00)00115-x</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zpitjv3j2nb4npgg4kn3g63riu">fatcat:zpitjv3j2nb4npgg4kn3g63riu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20090106153557/http://rii.ricoh.com/~hull/pubs/lee_prl01.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/b2/43/b2436d03a31be7600278f251f997b77779d9d0e0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/s0167-8655(00)00115-x"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> elsevier.com </button> </a>

Substitution deciphering based on HMMs with applications to compressed document processing

Dar-Shyang Lee
<span title="">2002</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/3px634ph3vhrtmtuip6xznraqi" style="color: black;">IEEE Transactions on Pattern Analysis and Machine Intelligence</a> </i> &nbsp;
We show that our algorithm is more accurate than relaxation and much more robust in the presence of noise, making it useful for applications in compressed document processing.  ...  In this paper, we propose a new solution to substitution deciphering based on hidden Markov models.  ...  Three indexing tables, one for each language, were constructed using a variation of n-gram descriptors generated from the deciphered outputs [10] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpami.2002.1114860">doi:10.1109/tpami.2002.1114860</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/dwb36tm5wvamdblxeczwoam7cq">fatcat:dwb36tm5wvamdblxeczwoam7cq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20060520232350/http://rii.ricoh.com:80/~dsl/publications/dsl_pami1202.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/37/44/37447116f7f2d22d24aa01c866315c0d02b8150f.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpami.2002.1114860"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

On Extractive and Abstractive Neural Document Summarization with Transformer Language Models [article]

Sandeep Subramanian, Raymond Li, Jonathan Pilault, Christopher Pal
<span title="2020-04-28">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We perform a simple extractive step before generating a summary, which is then used to condition the transformer language model on relevant information before being tasked with generating a summary.  ...  We present a method to produce abstractive summaries of long documents that exceed several thousand words via neural abstractive summarization.  ...  The decoder is in both cases is an LSTM. The extractive model considers the document as a list of N sentences D = (S 1 , . . . , S N ), and each sentence as a list of tokens.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.03186v2">arXiv:1909.03186v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4t33csrjzvdhlai2fkzw5e5hdi">fatcat:4t33csrjzvdhlai2fkzw5e5hdi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200506083342/https://arxiv.org/pdf/1909.03186v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.03186v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Adding linguistic constraints to document image decoding: comparing the iterated complete path and stack algorithms

Kris Popat, Daniel H. Greene, Justin K. Romberg, Dan S. Bloomberg, Paul B. Kantor, Daniel P. Lopresti, Jiangying Zhou
<span title="2000-12-21">2000</span> <i title="SPIE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ttq7pbic7jboja6n4lpaxbsfuu" style="color: black;">Document Recognition and Retrieval VIII</a> </i> &nbsp;
Beginning with an observed document image and a model of how the image has been degraded, Document Image Decoding recognizes printed text by attempting to find a most probable path through a hypothesized  ...  For methodological reasons, we continue to exercise tight control over the manner in which the images are produced and corrupted.  ...  Document Image Decoding We briefly review the essential elements of traditional Document Image Decoding.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.410844">doi:10.1117/12.410844</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/drr/PopatGRB01.html">dblp:conf/drr/PopatGRB01</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hq77qflv2vdytnckz3wsf4mjw4">fatcat:hq77qflv2vdytnckz3wsf4mjw4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20041028064024/http://www2.parc.com:80/istl/groups/gir/papers/Popat2001_stack.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/86/a0/86a045774daffa19bad44937ddc99f48ed4451bc.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.410844"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Improving Abstraction in Text Summarization

Wojciech Kryściński, Romain Paulus, Caiming Xiong, Richard Socher
<span title="">2018</span> <i title="Association for Computational Linguistics"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/u3ideoxy4fghvbsstiknuweth4" style="color: black;">Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing</a> </i> &nbsp;
First, we decompose the decoder into a contextual network that retrieves relevant parts of the source document, and a pretrained language model that incorporates prior knowledge about language generation  ...  Our model achieves results comparable to state-of-the-art models, as determined by ROUGE scores and human evaluations, while achieving a significantly higher level of abstraction as measured by n-gram  ...  Conclusions We introduced a new abstractive summarization model which uses an external language model in the decoder, as well as a new reinforcement learning reward to encourage summary abstraction.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/d18-1207">doi:10.18653/v1/d18-1207</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/emnlp/KryscinskiPXS18.html">dblp:conf/emnlp/KryscinskiPXS18</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5hmv4qlbhjfvxdq5vbkg4uesoe">fatcat:5hmv4qlbhjfvxdq5vbkg4uesoe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200508153947/https://www.aclweb.org/anthology/D18-1207.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/65/3d/653d2a29f9f7e67327758b249eb95f6cb9d14902.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/d18-1207"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Improving Abstraction in Text Summarization [article]

Wojciech Kryściński, Romain Paulus, Caiming Xiong, Richard Socher
<span title="2018-08-23">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
First, we decompose the decoder into a contextual network that retrieves relevant parts of the source document, and a pretrained language model that incorporates prior knowledge about language generation  ...  Our model achieves results comparable to state-of-the-art models, as determined by ROUGE scores and human evaluations, while achieving a significantly higher level of abstraction as measured by n-gram  ...  Conclusions We introduced a new abstractive summarization model which uses an external language model in the decoder, as well as a new reinforcement learning reward to encourage summary abstraction.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1808.07913v1">arXiv:1808.07913v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jjw2ouhs4vao5lraufx2m2z56e">fatcat:jjw2ouhs4vao5lraufx2m2z56e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200829005152/https://arxiv.org/pdf/1808.07913v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/33/56/335667621737bdbe03f1e654f412c626770edde8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1808.07913v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Construction of language models for an handwritten mail reading system

Olivier Morillot, Laurence Likforman-Sulem, Emmanuèle Grosicki, Christian Viard-Gaudin, Richard Zanibbi
<span title="2012-01-22">2012</span> <i title="SPIE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ttq7pbic7jboja6n4lpaxbsfuu" style="color: black;">Document Recognition and Retrieval XIX</a> </i> &nbsp;
To take into account linguistic information and enhance performance, a language model is introduced. This language model is based on bigrams and built from training document transcriptions only.  ...  This paper presents a system for the recognition of unconstrained handwritten mails. The main part of this system is an HMM recognizer which uses trigraphs to model contextual information.  ...  ACKNOWLEDGMENTS The authors would like to thank A-L Bianne-Bernard from Télécom ParisTech and Gilles Adda from LIMSI for fruitful discussions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.911965">doi:10.1117/12.911965</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/drr/MorillotLG12.html">dblp:conf/drr/MorillotLG12</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ug66hd42zjbanidwyc4fuvgsui">fatcat:ug66hd42zjbanidwyc4fuvgsui</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190503082302/https://hal.archives-ouvertes.fr/hal-00737407/document" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/5c/0d/5c0d04c6f6768f220feb5cab14eeb6b1439d2d37.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1117/12.911965"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Automatic Documentation of ICD Codes with Far-Field Speech Recognition [article]

Albert Haque, Corinna Fukushima
<span title="2018-11-27">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
As the standard language for diagnoses and billing, ICD codes serve as the foundation for medical documentation worldwide.  ...  By sampling from an unsupervised medical language model, our method is able to outperform existing methods.  ...  Language Model. Our language model is an n-gram model.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.11046v4">arXiv:1804.11046v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/t3te6tk6u5eqneb6khpxasze74">fatcat:t3te6tk6u5eqneb6khpxasze74</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191013071657/https://arxiv.org/pdf/1804.11046v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/6e/79/6e79a553dd126b182f03944c8f9204d2fde996f4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.11046v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

The A2iA Arabic Handwritten Text Recognition System at the Open HaRT2013 Evaluation

Theodore Bluche, Jerome Louradour, Maxime Knibbe, Bastien Moysset, Mohamed Faouzi Benzeghiba, Christopher Kermorvant
<span title="">2014</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ppbqu27wqffgnf3lfj4l4u6ycm" style="color: black;">2014 11th IAPR International Workshop on Document Analysis Systems</a> </i> &nbsp;
Large vocabulary selection techniques and ngram language modeling were used to provide a full paragraph recognition, without explicit word segmentation.  ...  These systems were based on an optical model using Long Short-Term Memory (LSTM) recurrent neural networks, trained to recognize the different forms of the Arabic characters directly from the image, without  ...  ACKNOWLEDGMENT This work was partly achieved as part of the Quaero Program, funded by OSEO, French State agency for innovation and was supported by the French Research Agency under the contract Cognilego  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/das.2014.40">doi:10.1109/das.2014.40</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/das/BlucheLKMBK14.html">dblp:conf/das/BlucheLKMBK14</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/axuddc7jlrejndr5x5zrdnrbfi">fatcat:axuddc7jlrejndr5x5zrdnrbfi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170810111324/http://www.a2ialab.com/lib/exe/fetch.php?media=publications:das2014_openhart.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/7d/b0/7db0acf523a55c3be8ab541f62434d690533aa40.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/das.2014.40"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 3,649 results