The AT&T spoken language understanding system

Spoken language understanding (SLU) aims at extracting meaning from natural language speech. Over the past decade, a variety of practical goal-oriented spoken dialog systems have been built for limited domains. SLU in these systems ranges from understanding predetermined phrases through fixed grammars, extracting some predefined named entities, extracting users' intents for call classification, to combinations of users' intents and named entities. In this paper, we present the SLU system of VoiceTone/spl reg/ (a service provided by AT&T where AT&T develops, deploys and hosts spoken dialog applications for enterprise customers). The SLU system includes extracting both intents and the named entities from the users' utterances. For intent determination, we use statistical classifiers trained from labeled data, and for named entity extraction we use rule-based fixed grammars. The focus of our work is to exploit data and to use machine learning techniques to create scalable SLU systems which can be quickly deployed for new domains with minimal human intervention. These objectives are achieved by 1) using the predicate-argument representation of semantic content of an utterance; 2) extending statistical classifiers to seamlessly integrate hand crafted classification rules with the rules learned from data; and 3) developing an active learning framework to minimize the human labeling effort for quickly building the classifier models and adapting them to changes. We present an evaluation of this system using two deployed applications of VoiceTone/spl reg/.     

智能仿生双向手语翻译系统

设计了一套智能仿生双向手语翻译系统,该系统主要由STM32微处理器、LD3320非特定语音识别模块、SYN6288语音合成芯片等组成,能够实现语音与手势的双向翻译。其中语音转手势部分可通过语音识别模块获得指令,手语机器人根据指令完成语音转动作的翻译。手势转语音部分通过数据手套捕获手臂的动作和姿态,识别手语动作,控制手语机器人发出语音。该系统具有成本低、识别度高、使用方便等优势,具有良好的应用前景。     

高斯超向量-支持向量机鉴别性语种识别系统

支持向量机在语种识别技术中获得了广泛的研究和应用,并且达到和传统混合高斯模型相当的性能。高斯超向量-支持向量机系统将高斯混合模型与支持向量机有效地结合起来,采用高斯超向量核函数,以支持向量机作为后端分类器。重点介绍基于高斯超向量-支持向量机的语种识别系统,并和传统的高斯混合模型系统进行比较。在美国国家标准技术研究院2003年和2007年语种识别评测数据集上进行实验。实验结果表明,高斯超向量-支持向量机系统相对于混合高斯模型建模的方法,在长时数据上有较明显的性能优势。     

集成化多媒体语言学习系统:系统结构与软件

针对现有各种类型多媒体语言学习系统多存在着操作界面层次较多等问题，介绍了一种类似“傻瓜”型的高性能价格比的集成式多媒体语言学习系统。     

可逆编程语言R-JAVA及其语言处理系统的设计

为了给基于可逆逻辑综合、可逆电路技术的可逆硬件平台设计可逆软件系统,提出基于JDK的可逆编程语言RJAVA.首先总结可逆编程语言的设计原则,然后为可逆语言R-JAVA设计文法规则和语言处理系统,其语言处理系统提取源程序中的可逆代码段并将其翻译成分别对应正反语义的等价JAVA代码,再将翻译后的JAVA代码交由JDK解释执行.最后通过示例程序表明R-JAVA源程序中的可逆函数既可正向运行,又可反向运行,从软件层次上实现了可逆计算.     

基于OMS可视化系统建模语言VSML的研究与开发

ＶＳＭＬ是集成化系统工程环境中用来进行应用系统建模的可视化系统建模语言,它采用图并茂的方式能从系统的功能、行为,数据和机构等方位来准确描述应用系统。介绍其底层数据对象管理系统ＯＭＳ和ＶＳＭＬ设想思想,根据其设计思想介绍其于ＭＯＳ的ＶＳＭＬ语言的开发与实现方法。     

The implementation of grammar in a hypermedia system for language learning

In this paper an interactive video system for language learning will be discussed. The various parts and media contained in the system are connected by a multitude of links which lend themselves to explorative learning. Particular attention is given to the way the grammar section is incorporated and how it is accessed.Signe Marie Sanne is associate professor in the Department of Romance Languages and Literatures, University of Bergen. Since 1989 she has developed interactive video systems for language learning in Italian, Norwegian as a second language, English, French and Spanish. Her research interests are related to instructional design.     

A portable language for system development

The design and implementation of a system development language which is economical to implement is described. Both macro processing and compiling techniques are used to translate it. The advantages and disadvantages of using such a language are discussed together with the main problems met in its implementation.     

The development of a specification language for a computer security system

The primary goal of this paper is to define an initial step towards the definition of ‘systems grammar’ based on the notion of formal languages which can be used as a ‘tool’ in the formal representation of computer security systems. Currently all modelling done on computer security systems is written up as mathematical models. These mathematical models are usually based on the mathematics of relations amongst objects, as opposed to the model described in this paper which is based on the theory of formal languages. This paper is aimed at people who are doing research on the logical aspects of computer security. It is the first of a series of two papers. This paper will give interim results and make more specific the definition of a ‘formal language’ which suits the computer security environment. The second paper will illustrate the actual use of the defined ‘formal language’ and show how to represent the characteristics of a computer security environment by using this ‘formal language’.     

Goedel语言类型系统

Goedel语言是一种说明性逻辑程序设计语言。该语言基于一阶逻辑，引入了一个多态多类的类型系统和多种新的语言成分，支持抽象数据类型和模块化程序设计等技术，语言本身也具有很强的说明性语义。详细介绍了Goedel语言的类型系统及其构造，对在逻辑程序设计语言中引入类型系统的作用进行了一些深入的分析和讨论。     

A Prolog-based natural language front-end system

A Prolog-based natural language front-end system is described with the following major issues of discussion: Domain independence of the syntax analyser was achieved by the ‘generate-and-test’ notion and the domain independent semantic representation; Determiners were treated as higher order predicates; A technique called ‘syntactic feature’ was employed to write a readable parser in Prolog.     

Green media-aware medical IoT system

Rapid proliferation in state-of-the art technologies has revolutionized the medical market for providing urgent, effective and economical health facilities to aging society. In this context media (i.e., video) transmission is considered as a quite significant step during first hour of the emergency for presenting a big and better picture of the event. However, the energy hungry media transmission process and slow progress in battery technologies have become a major and serious problem for the evolution of video technology in medical internet of things (MIoT) or internet of medical things (IoMT). So, promoting Green (i.e., energy-efficient) transmission during voluminous and variable bit rate (VBR) video in MIoT is a challenging and crucial problem for researchers and engineers. Therefore, the need arose to conduct research on Green media transmission techniques to cater the need of upcoming wearable healthcare devices. Thus, this research contributes in two distinct ways; first, a novel and sustainable Green Media Transmission Algorithm (GMTA) is proposed, second, a mathematical model and architecture of Green MIoT are designed by considering a 8-min medical media stream named, ‘Navigation to the Uterine Horn, transection of the horn and re-anastomosis’ to minimize transmission energy consumption in media-aware MIoT, and to develop feasible media transmission schedule for sensitive and urgent health information from physian to patients and vice vers through extremely power hungry natured wearable devices. The experimental results demonstrate that proposed GMTA saves energy up to 41%, to serve the community.