A methodology for turn-taking capabilities enhancement in Spoken Dialogue Systems using Reinforcement Learning. (January 2018)

• Record Type: Journal Article
• Authors: Khouzaimi, Hatim; Laroche, Romain; Lefèvre, Fabrice
• Is Part Of: Computer speech & language. Volume 47(2018)
• Page Start: 93
• Page End: 111
• Publication Date: 2018-01
• Subjects: Spoken Dialogue Systems -- Turn-taking -- Incremental dialogue -- Reinforcement Learning; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2017.07.006
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: We propose a new methodology to improve turn-taking capabilities in a Spoken Dialogue System using Reinforcement Learning. We describe a new incremental Dialogue System Architecture and a new incremental dialogue simulator. We train a Reinforcement Learning policy with the simulator. We show that it outperforms the handcrafted and the non-incremental baseline strategies. We validate these results in a live study with real users. Abstract: This article introduces a new methodology to enhance an existing traditional Spoken Dialogue System (SDS) with optimal turn-taking capabilities in order to increase dialogue efficiency. A new approach for transforming the traditional dialogue architecture into an incremental one at a low cost is presented: a new turn-taking decision module called the Scheduler is inserted between the Client and the Service. It is responsible for handling turn-taking decisions. Then, a User Simulator which is able to interact with the system using this new architecture has been implemented and used to train a new Reinforcement Learning turn-taking strategy. Compared to a non-incremental and a handcrafted incremental baselines, it is shown to perform better in simulation and in a real live experiment.