Towards Neural Speaker Modeling in Multi-Party Conversation: The Task, Dataset, and Models

AAAI 2018 pp. 8121-8122

doi:10.1609/AAAI.V32I1.12140 /aaai/2018/meng2018aaai-neural/

Abstract

In this paper, we address the problem of speaker classification in multi-party conversation, and collect massive data to facilitate research in this direction. We further investigate temporal-based and content-based models of speakers, and propose several hybrids of them. Experiments show that speaker classification is feasible, and that hybrid models outperform each single component.

PDF AAAI Semantic Scholar

Cite

Text

Meng et al. "Towards Neural Speaker Modeling in Multi-Party Conversation: The Task, Dataset, and Models." AAAI Conference on Artificial Intelligence, 2018. doi:10.1609/AAAI.V32I1.12140

Markdown

[Meng et al. "Towards Neural Speaker Modeling in Multi-Party Conversation: The Task, Dataset, and Models." AAAI Conference on Artificial Intelligence, 2018.](https://mlanthology.org/aaai/2018/meng2018aaai-neural/) doi:10.1609/AAAI.V32I1.12140

BibTeX

@inproceedings{meng2018aaai-neural,
  title     = {{Towards Neural Speaker Modeling in Multi-Party Conversation: The Task, Dataset, and Models}},
  author    = {Meng, Zhao and Mou, Lili and Jin, Zhi},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {8121-8122},
  doi       = {10.1609/AAAI.V32I1.12140},
  url       = {https://mlanthology.org/aaai/2018/meng2018aaai-neural/}
}