Using topic models for OCR correction

Despite several decades of research in document analysis, recognition of unconstrained handwritten documents is still considered a challenging task. Previous research in this area has shown that word recognizers perform adequately on constrained handwritten documents which typically use a restricted vocabulary (lexicon). But in the case of unconstrained handwritten documents, state-of-the-art word recognition accuracy is still below the acceptable limits. The objective of this research is to improve word recognition accuracy on unconstrained handwritten documents by applying a post-processing or OCR correction technique to the word recognition output. In this paper, we present two different methods for this purpose. First, we describe a lexicon reduction-based method by topic categorization of handwritten documents which is used to generate smaller topic-specific lexicons for improving the recognition accuracy. Second, we describe a method which uses topic-specific language models and a maximum-entropy based topic categorization model to refine the recognition output. We present the relative merits of each of these methods and report results on the publicly available IAM database.     

Morpho-syntactic post-processing of N-best lists for improved French automatic speech recognition

Many automatic speech recognition (ASR) systems rely on the sole pronunciation dictionaries and language models to take into account information about language. Implicitly, morphology and syntax are to a certain extent embedded in the language models but the richness of such linguistic knowledge is not exploited. This paper studies the use of morpho-syntactic (MS) information in a post-processing stage of an ASR system, by reordering N-best lists. Each sentence hypothesis is first part-of-speech tagged. A morpho-syntactic score is computed over the tag sequence with a long-span language model and combined to the acoustic and word-level language model scores. This new sentence-level score is finally used to rescore N-best lists by reranking or consensus. Experiments on a French broadcast news task show that morpho-syntactic knowledge improves the word error rate and confidence measures. In particular, it was observed that the errors corrected are not only agreement errors and errors on short grammatical words but also other errors on lexical words where the hypothesized lemma was modified.     

A two-pass approach for handling out-of-vocabulary words in a large vocabulary recognition task

This paper addresses the problem of recognizing a vocabulary of over 50,000 city names in a telephone access spoken dialogue system. We adopt a two-stage framework in which only major cities are represented in the first stage lexicon. We rely on an unknown word model encoded as a phone loop to detect OOV city names (referred to as ‘rare city’ names). We use SpeM, a tool that can extract words and word-initial cohorts from phone graphs from a large fallback lexicon, to provide an N-best list of promising city name hypotheses on the basis of the phone graph corresponding to the OOV. This N-best list is then inserted into the second stage lexicon for a subsequent recognition pass.Experiments were conducted on a set of spontaneous telephone-quality utterances; each containing one rare city name. It appeared that SpeM was able to include nearly 75% of the correct city names in an N-best hypothesis list of 3000 city names. With the names found by SpeM to extend the lexicon of the second stage recognizer, a word accuracy of 77.3% could be obtained. The best one-stage system yielded a word accuracy of 72.6%. The absolute number of correctly recognized rare city names almost doubled, from 62 for the best one-stage system to 102 for the best two-stage system. However, even the best two-stage system recognized only about one-third of the rare city names retrieved by SpeM. The paper discusses ways for improving the overall performance in the context of an application.     

Beyond ASR 1-best: Using word confusion networks in spoken language understanding

We are interested in the problem of robust understanding from noisy spontaneous speech input. With the advances in automated speech recognition (ASR), there has been increasing interest in spoken language understanding (SLU). A challenge in large vocabulary spoken language understanding is robustness to ASR errors. State of the art spoken language understanding relies on the best ASR hypotheses (ASR 1-best). In this paper, we propose methods for a tighter integration of ASR and SLU using word confusion networks (WCNs). WCNs obtained from ASR word graphs (lattices) provide a compact representation of multiple aligned ASR hypotheses along with word confidence scores, without compromising recognition accuracy. We present our work on exploiting WCNs instead of simply using ASR one-best hypotheses. In this work, we focus on the tasks of named entity detection and extraction and call classification in a spoken dialog system, although the idea is more general and applicable to other spoken language processing tasks. For named entity detection, we have improved the F-measure by using both word lattices and WCNs, 6–10% absolute. The processing of WCNs was 25 times faster than lattices, which is very important for real-life applications. For call classification, we have shown between 5% and 10% relative reduction in error rate using WCNs compared to ASR 1-best output.     

Finding consensus in speech recognition: word error minimization and other applications of confusion networks

We describe a new framework for distilling information from word lattices to improve the accuracy of the speech recognition output and obtain a more perspicuous representation of a set of alternative hypotheses. In the standard MAP decoding approach the recognizer outputs the string of words corresponding to the path with the highest posterior probability given the acoustics and a language model. However, even given optimal models, the MAP decoder does not necessarily minimize the commonly used performance metric, word error rate (WER). We describe a method for explicitly minimizing WER by extracting word hypotheses with the highest posterior probabilities from word lattices. We change the standard problem formulation by replacing global search over a large set of sentence hypotheses with local search over a small set of word candidates. In addition to improving the accuracy of the recognizer, our method produces a new representation of a set of candidate hypotheses that specifies the sequence of word-level confusions in a compact lattice format. We study the properties of confusion networks and examine their use for other tasks, such as lattice compression, word spotting, confidence annotation, and reevaluation of recognition hypotheses using higher-level knowledge sources.     

基于多模型融合的汉语介词短语识别

该文提出了一种多模型融合的介词短语识别方法,不仅能识别并列型介词短语,而且提高了嵌套型介词短语的识别精度。首先,利用简单名词短语识别模型识别出语料中的短语信息并进行融合,简化语料,降低介词短语内部复杂性;其次,用CRF模型识别嵌套的内层介词短语,即若存在嵌套则识别嵌套的内层,若无嵌套则识别该介词短语;最后,将初始语料中识别出来的内层介词短语进行分词融合并修改其特征信息,重新训练外层介词短语识别模型进行识别。在内外层介词短语自动识别后,利用双重错误校正系统对识别的介词短语进行校正。在2000年《人民日报》语料中的7 028个介词短语进行五倍交叉实验,结果表明,该方法识别的介词短语的正确率、召回率、F值分别为94.11%、94.02%、94.06%,比基于简单名词短语的介词短语识别方法(baseline)分别提高了1.09％、1.07％、1.08％,有效提高了介词短语识别的性能。     

N-gram posterior probability confidence measures for statistical machine translation: an empirical study

We report an empirical study of n-gram posterior probability confidence measures for statistical machine translation (SMT). We first describe an efficient and practical algorithm for rapidly computing n-gram posterior probabilities from large translation word lattices. These probabilities are shown to be a good predictor of whether or not the n-gram is found in human reference translations, motivating their use as a confidence measure for SMT. Comprehensive n-gram precision and word coverage measurements are presented for a variety of different language pairs, domains and conditions. We analyze the effect on reference precision of using single or multiple references, and compare the precision of posteriors computed from k-best lists to those computed over the full evidence space of the lattice. We also demonstrate improved confidence by combining multiple lattices in a multi-source translation framework.     

基于加权贝叶斯的脱机手写阿文单词识别

针对手写阿拉伯单词书写连笔,且相似词较多的特点,该文提出一种新的脱机手写文字识别算法。该算法以固定组件为成分拆分阿拉伯单词,构建自组件特征至单词类别的加权贝叶斯推理模型。算法结合单词组件分割、多级混合式组件识别、组件加权系数估计等,计算单词类别的后验概率并得到单词识别结果。在IFN/ENIT库上的实验,获得了90.03%的单词识别率,证实组件分解对笔画连写具有鲁棒性,组件识别能提高相似词的辨别能力,而且该算法所需训练类别少,易向大词汇量识别扩展。     

Neural network language models for off-line handwriting recognition

Unconstrained off-line continuous handwritten text recognition is a very challenging task which has been recently addressed by different promising techniques. This work presents our latest contribution to this task, integrating neural network language models in the decoding process of three state-of-the-art systems: one based on bidirectional recurrent neural networks, another based on hybrid hidden Markov models and, finally, a combination of both. Experimental results obtained on the IAM off-line database demonstrate that consistent word error rate reductions can be achieved with neural network language models when compared with statistical N-gram language models on the three tested systems. The best word error rate, 16.1%, reported with ROVER combination of systems using neural network language models significantly outperforms current benchmark results for the IAM database.     

Supporting electronic ink databases

The emergence of the pen as the main interface device for personal digital assistants and pen-computers has made handwritten text, and more generally ink, a first-class object. As for any other type of data, the need of retrieval is a prevailing one. Retrieval of handwritten text is more difficult than that of conventional data since it is necessary to identify a handwritten word given slightly different variations in its shape. The current way of addressing this is by using handwriting recognition, which is prone to errors and limits the expressiveness of ink. Alternatively, one can retrieve from the database handwritten words that are similar to a query handwritten word using techniques borrowed from pattern and speech recognition. In this paper, an indexing technique based on Hidden Markov Models is proposed. Its implementation and its performance is reported in this paper.     

Robust named entity detection from optical character recognition output

In this paper, we focus on information extraction from optical character recognition (OCR) output. Since the content from OCR inherently has many errors, we present robust algorithms for information extraction from OCR lattices instead of merely looking them up in the top-choice (1-best) OCR output. Specifically, we address the challenge of named entity detection in noisy OCR output and show that searching for named entities in the recognition lattice significantly improves detection accuracy over 1-best search. While lattice-based named entity (NE) detection improves NE recall from OCR output, there are two problems with this approach: (1) the number of false alarms can be prohibitive for certain applications and (2) lattice-based search is computationally more expensive than 1-best NE lookup. To mitigate the above challenges, we present techniques for reducing false alarms using confidence measures and for reducing the amount of computation involved in performing the NE search. Furthermore, to demonstrate that our techniques are applicable across multiple domains and languages, we experiment with optical character recognition systems for videotext in English and scanned handwritten text in Arabic.     

Confidence modeling for handwriting recognition: algorithms and applications

Confidence scoring can assist in determining how to use imperfect handwriting-recognition output. We explore a confidence-scoring framework for post-processing recognition for two purposes: deciding when to reject the recognizer's output, and detecting when to change recognition parameters e.g., to relax a word-set constraint. Varied confidence scores, including likelihood ratios and posterior probabilities, are applied to an Hidden-Markov-Model (HMM) based on-line recognizer. Receiver-operating characteristic curves reveal that we successfully reject 90% of word recognition errors while rejecting only 33% of correctly-recognized words. For isolated digit recognition, we achieve 90% correct rejection while limiting false rejection to 13%.     

Local reference lines for handwritten phrase recognition

Reference line information has been used for diverse purposes in handwriting research, including word case classification, OCR, and holistic word recognition. In this paper, we argue that the commonly used global reference lines are inadequate for many handwritten phrase recognition applications. Individual words may be written at different orientations or vertically displaced with respect to one another. A function used to approximate the implicit baseline will not be differentiable or even continuous at some points. We have presented the case for local reference lines and illustrate its successful use in a system that verifies street name phrases in a postal application.     

线性合成的双粒度 RNN 集成系统

针对脱机文字识别,提出了一种基于线性合成的双粒度递归神经网络(Recurrent neural net work, RNN)集成系统.首先,使用单词RNN对未知图 像进行识别;然后,依据识别结果进行字符分割,使用字符RNN对分割后的字符进行识别,并利用查表法计算字符的后验概率;最后,综合两个RNN的识别结果决定最终单词输出.在CAPTCHA识别 和手写识别上的实验结果证明了该系统的有效性.     

Use of adaptive segmentation in handwritten phrase recognition

Research in handwriting recognition has thus far been primarily focused on recognizing words and phrases. In fact, phrases are usually treated as a concatenation of the constituent words making it in essence an enhanced word recognizer. In this paper we present a methodology that will take advantage of the spacing between the words in a phrase to aid the recognition process. The novelty of our approach lies in the fact that the determination of word breaks is made in a manner that adapts to the writing style of the individual. The parameters that decide whether a particular gap between components is an inter-word gap or an inter-character gap are computed without the necessity of generalizing over a large training set. Rather, it is tuned to the distribution of the gaps within the instance of the phrase image being examined. We compare our approach to the methods described in the literature that simply ignore the significance of gaps in a phrase. Our experiments show an improvement of about 5% in recognition rates. On a test set of about 1400 phrase images the segmentation method “misses” only 2% of the true word break points.     

Character confidence based on N-best list for keyword spotting in online Chinese handwritten documents

In keyword spotting from handwritten documents by text query, the word similarity is usually computed by combining character similarities, which are desired to approximate the logarithm of the character probabilities. In this paper, we propose to directly estimate the posterior probability (also called confidence) of candidate characters based on the N-best paths from the candidate segmentation-recognition lattice. On evaluating the candidate segmentation-recognition paths by combining multiple contexts, the scores of the N-best paths are transformed to posterior probabilities using soft-max. The parameter of soft-max (confidence parameter) is estimated from the character confusion network, which is constructed by aligning different paths using a string matching algorithm. The posterior probability of a candidate character is the summation of the probabilities of the paths that pass through the candidate character. We compare the proposed posterior probability estimation method with some reference methods including the word confidence measure and the text line recognition method. Experimental results of keyword spotting on a large database CASIA-OLHWDB of unconstrained online Chinese handwriting demonstrate the effectiveness of the proposed method.     

使用无监督网络MLLR自适应改进算法的语音识别

介绍了一种基于词网的最大似然线性回归(Lattice-MLLR)无监督自适应算法,并进行了改进。Lattice-MLLR是根据解码得到的词网估计MLLR变换参数,词网的潜在误识率远小于识别结果,因此可以使参数估计更为准确。Lattice-MLLR的一个很大缺点是计算量极大,较难实用,对此本文提出了两个改进技术:(1)利用后验概率压缩词网;(2)利用单词的时间信息限制状态统计量的计算范围。实验测定Lattice-MLLR的误识率比传统MLLR相对下降了3.5%,改进技术使Lattice-MLLR计算量下降幅度超过了87.9%。     

Level of interest sensing in spoken dialog using decision-level fusion of acoustic and lexical evidence

Automatic detection of a user's interest in spoken dialog plays an important role in many applications, such as tutoring systems and customer service systems. In this study, we propose a decision-level fusion approach using acoustic and lexical information to accurately sense a user's interest at the utterance level. Our system consists of three parts: acoustic/prosodic model, lexical model, and a model that combines their decisions for the final output. We use two different regression algorithms to complement each other for the acoustic model. For lexical information, in addition to the bag-of-words model, we propose new features including a level-of-interest value for each word, length information using the number of words, estimated speaking rate, silence in the utterance, and similarity with other utterances. We also investigate the effectiveness of using more automatic speech recognition (ASR) hypotheses (n-best lists) to extract lexical features. The outputs from the acoustic and lexical models are combined at the decision level. Our experiments show that combining acoustic evidence with lexical information improves level-of-interest detection performance, even when lexical features are extracted from ASR output with high word error rate.