Effective Context Modeling Framework for Emotion Recognition in Conversations

Table of Contents

• Introduction
• Installation
• Usage
• Acknowledgement
• Citation

Introduction

This is the official implementation of the paper "Effective Context Modeling Framework for Emotion Recognition in Conversations". Our paper is published at ICASSP 2025 🎉.

Figure: Detailed architecture of (A) the proposed ConxGNN, (B) Inception Graph Block, and (C) HyperBlock.

Installation

Install the dependencies:

 conda env create -f environment/environment.yml

Read environment/helper.txt if some libraries can't be installed.

Usage

To train the model, run the following command:

python train.py configs/meld.yaml       # for MELD
python train.py configs/iemocap6.yaml   # for IEMOCAP

Acknowledgement

Part of the code is borrowed from the following repositories. We would like to thank the authors for their great work.

• MultiEMO
• M3Net
• CORECT

Citation

If you find this work helpful, please consider citing our paper:

@INPROCEEDINGS{10888112,
  author={Van, Cuong Tran and Tran, Thanh V. T. and Nguyen, Van and Hy, Truong Son},
  booktitle={ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, 
  title={Effective Context Modeling Framework for Emotion Recognition in Conversations}, 
  year={2025},
  volume={},
  number={},
  pages={1-5},
  keywords={Emotion recognition;Attention mechanisms;Limiting;Speech recognition;Oral communication;Benchmark testing;Graph neural networks;Data models;Speech processing;Context modeling;Emotion Recognition in Conversations;Graph Neural Network;Hypergraph;Multimodal},
  doi={10.1109/ICASSP49660.2025.10888112}}