TAIL: Text-Audio Incremental Learning

Yingfei Sun, Xu Gu, Wei Ji +3

Many studies combine text and audio to capture multi-modal information but they overlook the model's generalization ability on new datasets. Introducing new datasets may affect the…