检索结果-内蒙古大学图书馆

Automatic recognition and understanding of spoken language - A first step toward natural human-machine communication

引用

PROCEEDINGS OF THE IEEE 2000年第8期88卷 1142-1165页

作者： Juang, BH Furui, S Lucent Technol Inc Murray Hill NJ 07974 USA Tokyo Inst Technol Tokyo 1528552 Japan

The promise of a powerful computing device to help people in productivity as well as in recreation can only be realized with proper human-machine communication. Automatic recognition and understanding of spoken language is the first and probably the most important step toward natural human-machine interaction. Research in this fascinating field in the past few decades has produced remarkable results, leading to many exciting expectations as well as new challenges. In this paper, we summarize the development of the spoken language technology from both a vertical (the chronology) and a horizontal (the spectrum of technical approaches) perspective. We highlighted the introduction of statistical methods in dealing with language-related problems as it represents a paradigm shift in the research field of spoken language processing. statistical methods are designed to allow the machine to learn, directly from data, structure regularities in the speech signal for the purpose of automatic speech recognition and understanding. Today, research results in spoken language processing have led to a number of successful applications ranging from dictation software for personal computers and telephone-call processing systems for automatic call routing to automatic subcaptioning for television broadcast. We analyze the technical successes that support these applications. Along with an assessment of the state-of-the-art in this board technical field, we also discuss the limitations of the current technology can be presented as the basis to inspire future advances.

关键词： acoustic modeling acoustic-phonetics articulation automatic recognition and understanding Bayes risk cepstral distance continuous speech recognition detection-based approach dialogue systems discriminative training dynamic programming finite state machine forward-backward algorithm generalized phone models grammar hidden Markov models human-machine communication isolated word recognition language modeling language structure linear prediction maximum a posteriori maximum-likelihood estimation noise perplexity probability distribution of speech pronunciation modeling robustness search algorithms short-time spectral analysis signal analysis speech dictation speech distortion speech representations spoken language processing technology statistical language processing statistical pattern recognition

来源：评论

学校读者我要写书评

暂无评论

The interaction of top-down and bottom-up statistics in the resolution of syntactic category ambiguity

引用

JOURNAL OF MEMORY AND language 2006年第3期54卷 363-388页

作者： Gibson, E MIT Dept Brain & Cognit Sci Cambridge MA 02139 USA

This paper investigates how people resolve syntactic category ambiguities when comprehending sentences. It is proposed that people combine: (a) context-dependent syntactic expectations (top-down statistical information) and (b) context-independent lexical-category frequencies of words (bottom-up statistical information) in order to resolve ambiguities in the lexical categories of words. Three self-paced reading experiments were conducted involving the ambiguous word "that" in different syntactic environments in order to test these and other hypotheses. The data support the topdown/bottom-up approach in which the relative frequencies of lexical entries for a word are tabulated independent of context. Data from other experiments from the literature are discussed with respect to the model proposed here. (c) 2005 Elsevier Inc. All rights reserved.

关键词： sentence comprehension synactic category disambiguation statistical language processing frequency

来源：评论

学校读者我要写书评

暂无评论

Answering subcognitive Turing Test questions: a reply to French

引用

JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE 2001年第4期13卷 409-419页

作者： Turney, PD Natl Res Council Canada Inst Informat Technol Ottawa ON K1A 0R6 Canada

Robert French has argued that a disembodied computer is incapable of passing a Turing Test that includes subcognitive questions. Subcognitive questions are designed to probe the network of cultural and perceptual associations that humans naturally develop as we live, embodied and embedded in the world. This paper shows how it is possible for a disembodied computer to answer subcognitive questions appropriately, contrary to French's claim. The paper's approach to answering subcognitive questions is to use statistical information extracted from a very large collection of text. In particular, it is shown how it is possible to answer a sample of subcognitive questions taken from French, by issuing queries to a search engine that indexes about 350 million Web pages. This simple algorithm may shed light on the nature of human (sub-) cognition, but the scope of this paper is limited to demonstrating that French is mistaken: a disembodied computer can answer subcognitive questions.

关键词： Turing Test subcognitive questions subcognition mutual information lexical analysis word association statistical language processing web mining PMI-IR

来源：评论

学校读者我要写书评

暂无评论

Romanization of Thai Proper Names Based on Popularity of Usages

Romanization of Thai Proper Names Based on Popularity of Usa...

引用

13th Pacific-Asia Conference on Knowledge and Data Mining

作者： Tangverapong, Akegapon Suchato, Atiwong Punyabukkana, Proadpran Chulalongkorn Univ Fac Engn Dept Comp Engn Spoken Language Syst Res Grp Bangkok 10330 Thailand

ISBN: (纸本)9783642013065

The lack of standards for Romanization of Thai proper names makes searching activity a challenging task. This is particularly important when searching for people-related documents based on orthographic representation of their names using either solely Thai or English alphabets. Romanization based directly on the names' pronunciations often fails to deliver exact English spellings due to the non-1-to-1 mapping from Thai to English spelling and personal preferences. This paper proposes a Romanization approach where popularity of usages is taken into consideration. Thai names are parsed into sequences of grams, units of syllable-sized or larger governed by pronunciation and spelling constraints in both Thai and English writing systems. A Gram lexicon is constructed from a corpus of more than 130,000 names. statistical models are trained accordingly based on the Gram lexicon. The proposed method significantly outperformed the current Romanization approach. Approximately 46% to 75% of the correct English spellings are covered when the number of proposed hypotheses increases from 1 to 15.

关键词： Thai Romanization statistical language processing Machine Translation

来源：评论

学校读者我要写书评

暂无评论

Anatomy of Building Marathi N-Grams

Anatomy of Building Marathi N-Grams

引用

International Conference on Computing, Analytics and Security Trends (CAST)

作者： Gajendragadkar, Uma Joshi, Sarang SPPU COEP Comp Dept Pune Maharashtra India SPPU PICT Comp Dept Pune Maharashtra India

ISBN: (纸本)9781509013388

statistical language processing uses n-grams. Building n-grams from a corpus is not a trivial process. It requires careful selection of the sources that make the corpus, data collection over a long period, data cleansing, script mix removal and finally building usable n-grams for a given language. While English language n-grams are available from Google and are very popular, most of the Indic languages do not have such readily available corpus. This paper describes building n-gram corpus for Marathi, a language used daily by 70 million people in India.

关键词： Marathi n-gram statistical language processing

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：