Generative Encoder-Decoder Models for Task-Oriented Spoken Dialog Systems with Chatting Capability
This work addresses the problem of building more flexible and domain-general task-oriented dialog systems for applications like customer service, though it is incremental in extending existing models to new tasks.
The paper tackled the challenge of applying generative encoder-decoder models to task-oriented dialog systems by introducing a framework that enables slot-value independent decision-making and database interaction, achieving good performance in offline metrics and human task success rates.
Generative encoder-decoder models offer great promise in developing domain-general dialog systems. However, they have mainly been applied to open-domain conversations. This paper presents a practical and novel framework for building task-oriented dialog systems based on encoder-decoder models. This framework enables encoder-decoder models to accomplish slot-value independent decision-making and interact with external databases. Moreover, this paper shows the flexibility of the proposed method by interleaving chatting capability with a slot-filling system for better out-of-domain recovery. The models were trained on both real-user data from a bus information system and human-human chat data. Results show that the proposed framework achieves good performance in both offline evaluation metrics and in task success rate with human users.