The utterance selection module is a component of a dialog system that is responsible for selecting the most appropriate utterance or response to a given user input. This module typically receives input from a variety of other modules, such as the natural language understanding module, the conversation control module, and the ontology extraction module. It then uses this information to select the most appropriate utterance or response from a set of pre-defined options.
The utterance selection module typically includes a set of rules or algorithms that define how the system should select the most appropriate utterance or response. For example, the module may use rules to identify the key words and phrases in the user’s input, and then use these words and phrases to select an appropriate response from a set of pre-defined options. It may also use algorithms to evaluate the relevance or likelihood of different responses, and then select the one that is most appropriate based on the current context and the user’s goals and preferences.
Once the utterance has been selected, it is passed to the language generation module, which converts it into natural language text that can be output to the user. The language generation module may also apply additional processing, such as prosody modeling or voice transformation, to make the selected utterance more natural and expressive.
Overall, the function of the utterance selection module is to help the dialog system select the most appropriate utterance or response to a given user input. This can help improve the quality and accuracy of the system’s responses, and make the interaction more natural and effective for the user.
Summarization of spontaneous conversations [PDF] from utoronto.ca X Zhu… – Ninth International Conference on Spoken …, 2006 – isca-speech.org … spontaneous speech, our summarizer copes with them. Instead of removing them immediately as in , disfluency information is fed into the utterance selection module together with other features. Later, if the summaries are presented … Cited by 32 – Related articles – All 4 versions
[PDF] A critical reassessment of evaluation baselines for speech summarization [PDF] from aclweb.org G Penn… – Proceedings of ACL-HLT. Columbus, OH, 2008 – aclweb.org … To obtain a trainable utterance selection module that can utilize and compare rich features, we formu- lated utterance selection as a standard binary clas- sification problem, and experimented with several state-of-the-art classifiers, including linear discrim- inant analysis LDA … Cited by 25 – Related articles – View as HTML – All 13 versions
Remote education based on robot edutainment A Yorita, T Hashimoto, H Kobayashi… – Progress in Robotics, 2009 – Springer … lecture mode. The proposed conversation system used in the interaction mode is composed of three interrelated modules; (1) topic selection modules, (2) conversation control mod- ule, and (3) utterance selection module. The … Cited by 3 – Related articles – All 3 versions
Summarizing Spoken Documents Through Utterance Selection [PDF] from utoronto.ca X Zhu – 2010 – tspace.library.utoronto.ca Page 1. SUMMARIZING SPOKEN DOCUMENTS THROUGH UTTERANCE SELECTION by Xiaodan Zhu A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Computer Science University of Toronto … Cited by 2 – Related articles – All 3 versions
Conversation system based on Boltzmann selection and Bayesian networks for a partner robot N Kubota… – … Communication, 2009. RO-MAN 2009. The …, 2009 – ieeexplore.ieee.org … This paper proposes a conversation system composed of topic selection module, conversation control module and utterance selection module. … We apply conversation system composed of topic selection modules, conversation control module and utterance selection module. … Related articles
Conversation system for robot partners based on informationally structured space N Kubota, T Mori… – Robotic Intelligence In …, 2011 – ieeexplore.ieee.org … We apply conversation system composed of topic selection modules, conversation control module and utterance selection module. … The daily conversation mode is performed by conversation control module and utterance selection module. … Related articles
[PDF] Ecologically Valid Evaluation of Speech Summarization [PDF] from utoronto.ca A McCallum, C Munteanu, G Penn… – cs.utoronto.ca … A binary logistic regression classifier is used to train an utterance selection module that can make use of various lexical (MMR score, utterance length, etc.), structural (utterance position, etc.), and acoustic (pitch, energy, speaking rate, etc.) features, among others. … Related articles – View as HTML – All 2 versions
[DOC] Improving Speech Recognizer Performance in a Dialog System Using [DOC] from cmu.edu A Chotimongkol – cs.cmu.edu … The reranked concept error rate is 9.68%, which is a 1.22% relative improvement from the baseline. Adding an utterance selection module into a reranking process did not improve the reranking performance beyond the number achieved by reranking every utterance. … Related articles – View as HTML – All 3 versions