This paper discusses methods enhancing the selection of a “best” parsing tree from the output of natural language syntactic analysis. It presents a method for cutting away redundant parse trees based on the information obtained from a dependency tree-bank corpus. The effectivity of the enhanced parser is demonstrated by results of intersystem parser comparison. The test were run on the standard...

The paper presents a series of experiments on speech utterance classification performed on the ATIS corpus. We compare the performance of n-gram classifiers with that of Naive Bayes and maximum entropy classifiers. The n-gram classifiers have the advantage that one can use a single pass system (concurrent speech recognition and classification) whereas for Naive Bayes or maximum entropy classifi...

Internet-based Advanced Traveler Information Services (ATIS) provide the urban traveler with estimated travel times based on current roadway congestion. Survey research indicates that the vast majority of current ATIS users are satisfied consumers who feel they save time by utilizing these services on a regular basis. However, in numerous field experiments and simulation studies, ATIS users exp...

A potential benefit of Advanced Traveler Information Systems (ATIS) is the reduction of unexpected traffic delays. Non-recurrent congestion occurs haphazardly at any moment of the day. It may be a consequence of external changes in the demand for travel (special events, holidays, emergency evacuations, etc.) or in the capacities of the infrastructure (impact of weather conditions, accidents, ro...

Enhanced RFID tag technology especially in the UHF frequency range provides extended functionality like high operating range and sensing and monitoring capabilities. Such functionality requiring extended system structures including data acquisition units, real time clocks and active transmitters causes a high energy consumption of the tag and requires an on board energy store (battery). As a ke...

This paper proposes a method for incrementally translating English spoken language into Japanese. To realize simultaneous translation between languages with different word order, such as English and Japanese, our method utilizes the feature that the word order of a target language is flexible. To resolve the problem of generating a grammatically incorrect sentence, our method uses dependency st...

Internet Protocol-based television (IPTV) concerns video entertainment and represents a solution for interactive television-like services over IP-based networks. Operators and vendors are currently working on IPTV standardization efforts (e.g., ATIS/IIF, ITU-T IPTV-GSI, ETSI TISPAN) to bear wider availability and interoperability of IPTV as a secure, reliable, managed multimedia service. In thi...

This paper describes our complete results for improved lexical llers as well as two new kinds of llers, gives their results in unlimited speech recognition as well as for keyword spotting and compares them to the acoustic-phonetic ller in the case of keyword spotting. Tests have been conducted on di erent vocabularies derived from ATIS and the Wall Street Journal database. Results for keyword s...

Training of continuous density hidden Markov models (CDHMMs) is usually time-consuming and tedious due to the large number of model parameters involved. Recently we proposed a new derivative of CDHMM, the subspace distribution clustering hidden Markov model (SDCHMM) which tie CDHMMs at the ner level of subspace distributions, resulting in many fewer model parameters. An SDCHMM training algorith...

The contents of this report reflect the views of the authors who are responsible for the facts and the accuracy of the data presented herein. The contents do no necessarily reflect the official views or policies of the State of California. This report does not constitute a standard, specification, or regulation. Executive Summary The Bay Area ATIS (Advanced Traveler Information System) testbed ...