Dialogue Discourse-Aware Graph Model and Data Augmentation for Meeting Summarization
This work addresses the problem of improving meeting summarization accuracy for researchers and practitioners by introducing discourse-aware modeling and data augmentation, which is an incremental improvement.
This paper tackles the challenge of meeting summarization, which is difficult due to dynamic speaker interaction and limited training data. The authors propose a Dialogue Discourse-Aware Meeting Summarizer (DDAMS) that explicitly models discourse relations between utterances using a relational graph encoder. They also introduce a Dialogue Discourse-Aware Data Augmentation (DDADA) strategy, which expands the training corpus 20-fold and helps achieve state-of-the-art performance on the AMI and ICSI meeting datasets.
Meeting summarization is a challenging task due to its dynamic interaction nature among multiple speakers and lack of sufficient training data. Existing methods view the meeting as a linear sequence of utterances while ignoring the diverse relations between each utterance. Besides, the limited labeled data further hinders the ability of data-hungry neural models. In this paper, we try to mitigate the above challenges by introducing dialogue-discourse relations. First, we present a Dialogue Discourse-Dware Meeting Summarizer (DDAMS) to explicitly model the interaction between utterances in a meeting by modeling different discourse relations. The core module is a relational graph encoder, where the utterances and discourse relations are modeled in a graph interaction manner. Moreover, we devise a Dialogue Discourse-Aware Data Augmentation (DDADA) strategy to construct a pseudo-summarization corpus from existing input meetings, which is 20 times larger than the original dataset and can be used to pretrain DDAMS. Experimental results on AMI and ICSI meeting datasets show that our full system can achieve SOTA performance. Our codes will be available at: https://github.com/xcfcode/DDAMS.