43199

Автор(ы): 

Автор(ов): 

2

Параметры публикации

Тип публикации: 

Доклад

Название: 

Probabilistic methods for comparative analysis and optimization of scenarios and algorithms for managing speech dialogue

ISBN/ISSN: 

978-5-209-08291-0

Наименование конференции: 

  • International Conference on Analytical and Computational Methods in Probability Theory and its Applications (ACMPT-2017, Moscow)

Наименование источника: 

  • Proceedings of the International Conference on Analytical and Computational Methods in Probability Theory and its Applications (ACMPT-2017)

Город: 

  • Москва

Издательство: 

  • РУДН

Год издания: 

2017

Страницы: 

226-233
Аннотация
In the article, the basic characteristics of human-machine dialogue in speech dialogue systems are investigated by probabilistic methods: the probability of successful completion of a dialogue and evaluation of the duration of dialogues with various procedures for identifying and correcting errors. The analysis of various variants of dialogues is made and recommendations are received for optimizing the scenarios of the dialogues taking into account the reliability of the recognizer.

Библиографическая ссылка: 

Фархадов М.П., Петухова Н.В. Probabilistic methods for comparative analysis and optimization of scenarios and algorithms for managing speech dialogue / Proceedings of the International Conference on Analytical and Computational Methods in Probability Theory and its Applications (ACMPT-2017). М.: РУДН, 2017. С. 226-233.