Knowledge-grounded task-oriented dialogue modeling on spoken conversations track at DSTC10
A lot of recent work in dialogue modeling has been on written conversations, partly because of available data sets. However, written dialogues are not sufficient to fully capture the nature of spoken conversations as well as the potential effect of speech recognition errors on practical spoken dialogue systems. This challenge track aims to provide a new benchmark on spoken task-oriented conversations. We introduce the validation and test data sets for multi-domain dialogue state tracking and knowledge-grounded dialogue modeling tasks. The challenge track received a total of 99 entries from 21 participating teams in total. From the evaluation results, we observe that the data augmentation and model ensemble methods are two major factors that help enhance models’ generalization capabilities to unseen spoken data and achieve good performance on both tasks.