检索结果-内蒙古大学图书馆

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

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

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

Using random forests in the structured language model

Using random forests in the structured language model

引用

18th Annual Conference on Neural Information processing Systems, NIPS 2004

作者： Xu, Peng Jelinek, Frederick Department of Electrical and Computer Engineering Center for Language and Speech Processing Johns Hopkins University United States

ISBN: (纸本)0262195348

In this paper, we explore the use of Random Forests (RFs) in the structured language model (SLM), which uses rich syntactic information in predicting the next word based on words already seen. The goal in this work is to construct RFs by randomly growing Decision Trees (DTs) using syntactic information and investigate the performance of the SLM modeled by the RFs in automatic speech recognition. RFs, which were originally developed as classifiers, are a combination of decision tree classifiers. Each tree is grown based on random training data sampled independently and with the same distribution for all trees in the forest, and a random selection of possible questions at each node of the decision tree. Our approach extends the original idea of RFs to deal with the data sparseness problem encountered in language modeling. RFs have been studied in the context of n-gram language modeling and have been shown to generalize well to unseen data. We show in this paper that RFs using syntactic information can also achieve better performance in both perplexity (PPL) and word error rate (WER) in a large vocabulary speech recognition system, compared to a baseline that uses Kneser-Ney smoothing.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Robustness aspects of active learning for acoustic modeling 8

Robustness aspects of active learning for acoustic modeling

引用

8th International Conference on Spoken language processing, ICSLP 2004

作者： Kamm, Teresa M. Meyer, Gerard G.L. Center for Language and Speech Processing Department of Electrical and Computer Engineering Johns Hopkins University BaltimoreMD United States

We previously proposed [1] an iterative word-selective training method to cost-effectively utilize data preparation resources without compromising system performance. We continue this work and investigate the robustness of our active learning approach with respect to the starting conditions and further propose a stopping criterion that supports our objective to make effective use of transcription effort while minimizing system error. In particular, we demonstrate robustness to seven initial conditions, showing that we can select around 20 hours of training data and achieve a range of error rates between 8.6% and 9.0%, compared to an error rate of 10% when using all 50 hours of the training set. Additionally, we give empirical evidence that our proposed stopping criterion is in general a good predictor of when the minimum error rate is achieved, demonstrated for each of the initial conditions.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Using random forests in the structured language model 04

Using random forests in the structured language model

引用

Proceedings of the 18th International Conference on Neural Information processing Systems

作者： Peng Xu Frederick Jelinek Center for Language and Speech Processing Department of Electrical and Computer Engineering The Johns Hopkins University

In this paper, we explore the use of Random Forests (RFs) in the structured language model (SLM), which uses rich syntactic information in predicting the next word based on words already seen. The goal in this work is to construct RFs by randomly growing Decision Trees (DTs) using syntactic information and investigate the performance of the SLM modeled by the RFs in automatic speech ***, which were originally developed as classifiers, are a combination of decision tree classifiers. Each tree is grown based on random training data sampled independently and with the same distribution for all trees in the forest, and a random selection of possible questions at each node of the decision tree. Our approach extends the original idea of RFs to deal with the data sparseness problem encountered in language *** have been studied in the context of n-gram language modeling and have been shown to generalize well to unseen data. We show in this paper that RFs using syntactic information can also achieve better performance in both perplexity (PPL) and word error rate (WER) in a large vocabulary speech recognition system, compared to a baseline that uses Kneser-Ney smoothing.

关键词：

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：