We leverage different context windows when predicting the emotion of different utterances. New modules are included to realize variable-length context: 1) two speaker-aware units, which explicitly model inner- and inter-speaker dependencies to form distilled conversational context, and 2) a top-k normalization layer, which determines the most proper context windows from the conversational context to predict emotion. Experiments and ablation studies show that our approach outperforms several strong baselines on three public datasets.

翻译：我们针对不同话语的情感预测采用不同的上下文窗口，并引入新模块以实现可变长度上下文：1）两个说话人感知单元，分别显式建模说话人内部及跨说话人依赖关系，以形成精炼的对话上下文；2）一个top-k归一化层，用于从对话上下文中确定最合适的上下文窗口以预测情感。实验及消融研究表明，我们的方法在三个公开数据集上优于多个强基线模型。