Dialogue Efficiency Evaluation of Turn-Taking Phenomena in a Multi-layer Incremental Simulated Environment

Autor:	Fabrice Lefèvre, Romain Laroche, Hatim Khouzaimi
Rok vydání:	2015
Předmět:	Human–computer interaction Noise (signal processing) Computer science Process (engineering) Turn-taking Multi layer Utterance Simulation
Zdroj:	Communications in Computer and Information Science ISBN: 9783319213798 HCI (27)
DOI:	10.1007/978-3-319-21380-4_127
Popis:	We use a simulated environment for incremental dialogue to show that incremental processing offers new possibilities to make dialogue systems more robust to noise. Traditional dialogue systems wait until the end of the user’s utterance before processing it and they cannot be interrupted when taking the floor. On the contrary, incremental dialogue systems process the user’s speech signal on the flow, therefore, they are able to react quickly if an error is detected. First we show that mixed initiative strategies have the advantages of both system initiative and user initiative strategies and then we show that there is still room for improvement thanks to incremental processing.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::25c0112a7afb5a686f2e75705bc79b16 https://doi.org/10.1007/978-3-319-21380-4_127 Zobrazit plný text záznamu