检索结果-内蒙古大学图书馆

Information Theory and Applications Workshop, ITA

作者： Haolang Zhou Damianos Karakos Sanjeev Khudanpur Andreas G. Andreou Carey E. Priebe Department of Electrical and Computer Engineering and Center of Language and Speech Processing Johns Hopkins University Baltimore MD USA Department of Electrical and Computer Engineering Center for Language and Speech Processing Department of Applied Mathematics and Statistics Johns Hopkins University Baltimore MD USA

ISBN: (纸本)9781424439904

Generative statistical models with a very large number of parameters are frequently used in real-world data applications, such as large-vocabulary speech recognition (LVCSR). Complex models are needed in order to capture the ubiquitous variability in the observed signal, but data sparsity causes significant problems in their training. One way of dealing with data sparsity is to perform dimensionality reduction of the observed features, with the goal of reducing the model parameter space without sacrificing performance. When the data are Gaussian distributed, the dimensionality reduction can be done efficiently using the maximum likelihood criterion; this leads to the heteroscedastic linear discriminant analysis (HLDA), which is a natural extension of linear discriminant analysis (LDA) to the case where the class-conditional Gaussians have unequal covariance matrices. A further extension of HLDA to multiple transforms (MLDA) can also be tackled efficiently. This paper presents the theory behind HLDA and MLDA, and demonstrates their performance with synthetic data.

关键词： Gaussian distribution Linear discriminant analysis Covariance matrix speech recognition Maximum likelihood estimation Mathematics Acoustics Training data Mel frequency cepstral coefficient Distributed computing

来源：评论

学校读者我要写书评

暂无评论

Self-supervised discriminative training of statistical language models

Self-supervised discriminative training of statistical langu...

引用

IEEE Workshop on Automatic speech Recognition and Understanding

作者： Puyang Xu Damianos Karakos Sanjeev Khudanpur Department of Electrical and Computer Engineering Center for Language and Speech Processing USA Human Language Technology Center of Excellence Johns Hopkins University Baltimore MD USA Johns Hopkins University Baltimore MD US

ISBN: (纸本)9781424454785

A novel self-supervised discriminative training method for estimating language models for automatic speech recognition (ASR) is proposed. Unlike traditional discriminative training methods that require transcribed speech, only untranscribed speech and a large text corpus is required. An exponential form is assumed for the language model, as done in maximum entropy estimation, but the model is trained from the text using a discriminative criterion that targets word confusions actually witnessed in first-pass ASR output lattices. Specifically, model parameters are estimated to maximize the likelihood ratio between words w in the text corpus and w's cohorts in the test speech, i.e. other words that w competes with in the test lattices. Empirical results are presented to demonstrate statistically significant improvements over a 4-gram language model on a large vocabulary ASR task.

关键词： Automatic speech recognition Natural languages Lattices Maximum likelihood estimation Humans Testing speech recognition speech processing Entropy Parameter estimation

来源：评论

学校读者我要写书评

暂无评论

Exploiting prosodic breaks in language modeling with random forests

Exploiting prosodic breaks in language modeling with random ...

引用

4th International Conference on speech Prosody 2008, SP 2008

作者： Su, Yi Jelinek, Frederick Center for Language and Speech Processing Department of Electrical and Computer Engineering The Johns Hopkins University Baltimore MD United States

ISBN: (纸本)9780616220030

We propose a novel method of exploiting prosodic breaks in language modeling for automatic speech recognition (ASR) based on the random forest language model (RFLM), which is a collection of randomized decision tree language models and can potentially ask any questions about the history in order to predict the future. We demonstrate how questions about prosodic breaks can be easily incorporated into the RFLM and present two language models which treat prosodic breaks as observable and hidden variables, respectively. Meanwhile, we show empirically that a finer grained prosodic break is needed for language modeling. Experimental results showed that given prosodic breaks, we were able to reduce the LM perplexity by a significant margin, suggesting a prosodic N-best rescoring approach for ASR.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Sequential system combination for machine translation of speech

Sequential system combination for machine translation of spe...

引用

IEEE Spoken language Technology Workshop

作者： Damianos Karakos Sanjeev Khudanpur Center for Language and Speech Processing and Department of Electrical and Computer Engineering Johns Hopkins University Baltimore MD USA

System combination is a technique which has been shown to yield significant gains in speech recognition and machine translation. Most combination schemes perform an alignment between different system outputs in order to produce lattices (or confusion networks), from which a composite hypothesis is chosen, possibly with the help of a large language model. The benefit of this approach is two-fold: (i) whenever many systems agree with each other on a set of words, the combination output contains these words with high confidence; and (ii) whenever the systems disagree, the language model resolves the ambiguity based on the (probably correct) agreed upon context. The case of machine translation system combination is more challenging because of the different word orders of the translations: the alignment has to incorporate computationally expensive movements of word blocks. In this paper, we show how one can combine translation outputs efficiently, extending the incremental alignment procedure of (A-V.I. Rosti et al., 2008). A comparison between different system combination design choices is performed on an Arabic speech translation task.

关键词： speech processing speech recognition Automatic speech recognition Natural languages Lattices Context modeling Voting computer numerical control Contracts Sequences

来源：评论

学校读者我要写书评

暂无评论

Face verification algorithm based on combination of modular 2DPCA and CSLDA

引用

Jisuanji Yanjiu yu Fazhan/computer Research and Development 2008年第6期45卷 1029-1035页

作者： Yuan, Ning Wu, Xiaojun Wang, Shitong Yang, Jingyu Kittler, Josef School of Information Engineering Jiangnan University Wuxi 214122 China School of Computer Science and Technology Nanjing University of Science and Technology Nanjing 210094 China Center for Vision Speech and Signal Processing Department of Electrical Engineering University of Surrey Surrey GU2 7XH United Kingdom

An improved face verification algorithm is proposed based on the combination of modular 2DPCA and CSLDA in this paper. Feature extraction of client specific linear discriminant analysis (CSLDA) transforms an image matrix to a vector which causes great dimensionality and computational complexity. Furthermore, the local feature is not considered in CSLDA. Then the new method is studied to avoid the deficiency. The initial features are extracted with the original images which are divided into modular sub-images. The 2DPCA is performed to get the low dimensional features which can be computed conveniently. The local features are extracted efficiently using the proposed new method. Then CSLDA is utilized on the new pattern which is obtained through the modular 2DPCA to extract the final features. Compared with PCA, the discriminant information obtained from the between-class scatter matrix and within-class scatter matrix are included using CSLDA. Moreover, client specific subspace could describe the diversity of the different face better and has more robust discriminant information than the traditional LDA. The experimental results obtained on the facial database XM2VTS using the Lausanne protocol and the facial database ORL using the user-defined protocol show that the verification performance of the new method is superior to that of the primary method CSLDA.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Segmentation and alignment of parallel text for statistical machine translation

引用

Natural language engineering 2007年第3期13卷 235-260页

作者： Deng, Yonggang Kumar, Shankar Byrne, William Center for Language and Speech Processing Department of Electrical and Computer Engineering The Johns Hopkins University 3400 N. Charles St. Baltimore MD 21218 United States Google Inc. 1600 Amphitheatre Parkway Mountain View CA 94043 United States Department of Engineering Cambridge University Trumpington Street Cambridge CB2 1PZ United Kingdom

We address the problem of extracting bilingual chunk pairs from parallel text to create training sets for statistical machine translation. We formulate the problem in terms of a stochastic generative process over text translation pairs, and derive two *** alignment procedures based on the underlying alignment model. The first procedure is a now-standard dynamic programming alignment model which we use to generate an initial coarse alignment of the parallel text. The second procedure is a divisive clustering parallel text alignment procedure which we use to refine the first-pass alignments. This latter procedure is novel in that it permits the segmentation of the parallel text into sub-sentence units which are allowed to be reordered to improve the chunk alignment. The quality of chunk pairs are measured by the performance of machine translation systems trained from them. We show practical *** of divisive clustering as well as how system performance can be improved by exploiting portions of the parallel text that otherwise would have to be discarded. We also show that chunk alignment as a first step in word alignment can *** reduce word alignment error rate. © 2007 Cambridge University Press.

关键词： computer aided language translation

来源：评论

学校读者我要写书评

暂无评论

Unigram language models using diffusion smoothing over graphs

Unigram language models using diffusion smoothing over graph...

引用

2nd Workshop on Graph-Based Algorithms for Natural language processing, TextGraphs 2007

作者： Jedynak, Bruno Karakos, Damianos Dept. of Appl. Mathematics and Statistics Center for Imaging Sciences Johns Hopkins University Baltimore MD 21218-2686 United States Dept. of Electrical and Computer Engineering Center for Language and Speech Processing Johns Hopkins University Baltimore MD 21218-2686 United States

We propose to use graph-based diffusion techniques with data-dependent kernels to build unigram language models. Our approach entails building graphs, where each vertex corresponds uniquely to a word from a closed vocabulary, and the existence of an edge (with an appropriate weight) between two words indicates some form of similarity between them. In one of our constructions, we place an edge between two words if the number of times these words were seen in a training set differs by at most one count. This graph construction results in a similarity matrix with small intrinsic dimension, since words with the same counts have the same neighbors. Experimental results from a benchmark task from language modeling show that our method is competitive with the Good-Turing estimator.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Finding a Needle in a Haystack

Finding a Needle in a Haystack

引用

Annual Conference on Information Sciences and Systems (CISS)

作者： Bruno Jedynak Damianos Karakos Department of Applied Mathematics Johns Hopkins University MD USA Center for Language and Speech Processing and Department of Electrical and Computer Engineering Johns Hopkins University Baltimore MD USA

Summary form only given. We study a simplified version of the problem of target detectability in the presence of clutter. The target (the needle) is a sample of size N from a discrete distribution p. The clutter (the haystack) is made up of M independent samples of size JV from a distribution q (which is different from p, but with the same support). Two cases can be easily shown: (i) If M is fixed and JV goes to infinity, the target can be detected with probability that approaches 1. (ii) If TV is fixed and M goes to infinity, then, with probability approaching 1, the target cannot be detected. For the case where both JV, M go to infinity, we show that the asymptotic behavior of the optimal detector (if p, q are known) and of a plug-in detector (which estimates p, q on the fly) is determined by the asymptotic behavior of the quantity Mexp(-ND(p\\q)) : if it goes to zero (resp. infinity), then, with high probability, the target can (resp. cannot) be detected.

关键词： Needles H infinity control World Wide Web Detectors Mathematics Natural languages speech processing

来源：评论

学校读者我要写书评

暂无评论

A weighted finite state transducer translation template model for statistical machine translation

引用

Natural language engineering 2006年第1期12卷 35-75页

作者： Kumar, Shankar Deng, Yonggang Byrne, William Center for Language and Speech Processing Department of Electrical and Computer Engineering The Johns Hopkins University 3400 N. Charles St. Baltimore MD 21218 United States

We present a Weighted Finite State Transducer Translation Template Model for statistical machine translation. This is a source-channel model of translation inspired by the Alignment Template translation model. The model attempts to overcome the deficiencies of word-to-word translation models by considering phrases rather than words as units of translation. The approach we describe allows us to implement each constituent distribution of the model as a weighted finite state transducer or acceptor. We show that bitext word alignment and translation under the model can be performed with standard finite state machine operations involving these transducers. One of the benefits of using this framework is that it avoids the need to develop specialized search procedures, even for the generation of lattices or N-Best lists of bitext word alignments and translation hypotheses. We report and analyze bitext word alignment and translation performance on the Hansards French-English task and the FBIS Chinese-English task under the Alignment Error Rate, BLEU, NIST and Word Error-Rate metrics. These experiments identify the contribution of each of the model components to different aspects of alignment and translation performance. We finally discuss translation performance with large bitext training sets on the NIST 2004 Chinese-English and Arabic-English MT tasks. © 2005 Cambridge University Press.

关键词： computer aided language translation

来源：评论

学校读者我要写书评

暂无评论

Band-Independent Mask Estimation for Missing-Feature Reconstruction in the Presence of Unknown Background Noise

Band-Independent Mask Estimation for Missing-Feature Reconst...

引用

International Conference on Acoustics, speech, and Signal processing (ICASSP)

作者： Wooil Kim R.M. Stern Center for Robust Speech Systems Department of Electrical Eng. University of Technology Richardson TX USA Department of Electrical and Computer Engineering and Language Technologies Institute Carnegie Mellon University Pittsburgh PA USA

An effective mask estimation scheme for missing-feature reconstruction is described that achieves robust speech recognition in the presence of unknown noise. In previous work on Bayesian classification for mask estimation, white noise and colored noise were used for training mask estimators. This paper, which is concerned with both the simulation of a more diverse set of background environments and with mitigating the "sparse training" problem, describes a new Bayesian mask-estimation procedure in which each frequency band is trained independently. The new method employs colored noise for training, which is obtained by partitioning each frequency subband. We also propose a reevaluation method of voiced/unvoiced decisions to alleviate performance degradation that is caused by errors in pitch detection. Experimental results indicate that the proposed procedure in conjunction with cluster-based missing-feature imputation improves speech recognition accuracy on the Aurora 2.0 database in the presence for all types of background noise considered

关键词： Background noise speech recognition Bayesian methods Colored noise Noise robustness Working environment noise White noise Frequency diversity Degradation Databases

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：