This repository contains the dataset and code for our paper: MCM-CSD: MULTI-GRANULARITY CONTEXT MODELING WITH CONTRASTIVE SPEAKER DETECTION FOR EMOTION RECOGNITION IN REAL-TIME CONVERSATION
The complete code will be realeased after the review.
Install related dependencies:
pip install -r requirements.txtIn this paper, we focus on the text-based ERC, while some datasets are video resources. The original datasets can be found at IEMOCAP,MELD,DailyDialog and EmoryNLP.
Besides, we also provide the processed datasets(only in textual modality) that we use. You can download it from https://pan.baidu.com/s/18bp5mmQjYGLiRBs7HdNqzw. password：1mew
The complete code will be realeased after the review.