Spoken dialogue systems for traveller information
Spoken dialogue systems for traveller information
In recent years traveller information, including multimodal journey planning and real time information, is increasingly becoming available using consumer technology such as home computers, smartphones and PDAs. Recent UK government information initiatives, such as Transport Direct, seek to use such technologies to provide multimodal door-to-door traveller information. However, target users for such services will not always have access to suitable equipment to access this information, possibly due to socio-economic reasons, technology accessibility problems, or simply not having access to the technology on-trip. Manned telephone services, such as Traveline, accessible by fixed line or mobile telephone, fill this information gap, but suffer problems of financial cost, peak demand scalability, and out-of-hours availability. Such problems might be addressed by use of automating technologies, such as speech recognition.
This work presents research relevant to the provision of traveller information through Spoken Language Dialogue Systems (SLDS). It covers research on traveller information and user needs, the current state of SLDS technology, and describes the development of a research system designed to enable investigation of spoken dialogue phenomena, with a specific emphasis on the UK public transport information domain. A system pilot was conducted, as a first stage in an iterative development process, to provide some early feedback on the effectiveness of the design, as well as an opportunity to investigate relevant dialogue phenomena and assessment criteria, with a view to wider user group testing. Recommendations are made to the sponsoring company based on the research findings, as well as suggestions for future development of the experimental system.
University of Southampton
2006
MacKinnon, Grant N
(2006)
Spoken dialogue systems for traveller information.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
In recent years traveller information, including multimodal journey planning and real time information, is increasingly becoming available using consumer technology such as home computers, smartphones and PDAs. Recent UK government information initiatives, such as Transport Direct, seek to use such technologies to provide multimodal door-to-door traveller information. However, target users for such services will not always have access to suitable equipment to access this information, possibly due to socio-economic reasons, technology accessibility problems, or simply not having access to the technology on-trip. Manned telephone services, such as Traveline, accessible by fixed line or mobile telephone, fill this information gap, but suffer problems of financial cost, peak demand scalability, and out-of-hours availability. Such problems might be addressed by use of automating technologies, such as speech recognition.
This work presents research relevant to the provision of traveller information through Spoken Language Dialogue Systems (SLDS). It covers research on traveller information and user needs, the current state of SLDS technology, and describes the development of a research system designed to enable investigation of spoken dialogue phenomena, with a specific emphasis on the UK public transport information domain. A system pilot was conducted, as a first stage in an iterative development process, to provide some early feedback on the effectiveness of the design, as well as an opportunity to investigate relevant dialogue phenomena and assessment criteria, with a view to wider user group testing. Recommendations are made to the sponsoring company based on the research findings, as well as suggestions for future development of the experimental system.
This record has no associated files available for download.
More information
Published date: 2006
Identifiers
Local EPrints ID: 466152
URI: http://eprints.soton.ac.uk/id/eprint/466152
PURE UUID: 0e8c7373-8eb6-4c41-b9da-65176a5800ed
Catalogue record
Date deposited: 05 Jul 2022 04:32
Last modified: 05 Jul 2022 04:32
Export record
Contributors
Author:
Grant N MacKinnon
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics