dc.contributor.author | Narupiyakul, Lalita. | en_US |
dc.date.accessioned | 2014-10-21T12:33:49Z | |
dc.date.available | 2007 | |
dc.date.issued | 2007 | en_US |
dc.identifier.other | AAINR31502 | en_US |
dc.identifier.uri | http://hdl.handle.net/10222/54965 | |
dc.description | A speaker's utterance may convey different meanings to a hearer than what the speaker intended. Such ambiguities can be resolved by emphasizing accents at different positions. In human communication, the utterances are emphasized at a focus part to distinguish the important content and reduce ambiguity in the utterance. | en_US |
dc.description | In our Focus-to-Emphasize Tone (FET) system, we determine how the speaker's utterances are influenced by foci and speaker's intention. The relationships of focus information, speaker's intention and prosodic phenomena are investigated to recognize the intonation patterns and annotate the sentence with prosodic marks. The thesis consists of three parts: analysis, design and implementation, and evaluation of the FET system. The first part is the FET analysis. The relationships between focus, speaker's intention and prosody are analyzed. We consider how to define the intonation patterns using the speaker's intention and find which parts of the sentence serve as the focus parts. | en_US |
dc.description | In the second section, the design of the FET structure and subgrammar is developed using the information of focus, speaker's intention and prosody. Our FET structure and subgrammar are unification-based formalisms and can be used with the LKB system, which is an HPSG parsing system. The FET subgrammar includes typed constraints, a set of focus words, grammar rules, typed hierarchy, and typed feature structures for focus, speaker's intention and prosodic features. We implement the FET system as a proof-of-concept system, developed using the LKB system with our FET subgrammar. | en_US |
dc.description | The last part is the evaluation of the FET system including (i) the perceptual evaluation of the utterances conveying focus, and (ii) the evaluation of the prosodic annotation. The perceptual evaluation is performed by a listening test, in which participants must listen to different utterances of the same sentence and select a sound utterance that make the most sense from multiple choice questions in a dialogue. The CMU communicator dataset is used in the second evaluations and the results are discussed with respect to the performance of the FET system. | en_US |
dc.description | Thesis (Ph.D.)--Dalhousie University (Canada), 2007. | en_US |
dc.language | eng | en_US |
dc.publisher | Dalhousie University | en_US |
dc.publisher | | en_US |
dc.subject | Computer Science. | en_US |
dc.title | A unification-based focus system for prosodic analysis. | en_US |
dc.type | text | en_US |
dc.contributor.degree | Ph.D. | en_US |