Multimodal Target Speaker Extraction: Towards Unified Speaker Cues Across Modalities
A review of the current landscape and open problems in multimodal TSE from the perspective of auxiliary target cues drawn from multiple modalities and summarizes future directions concerning adaptive cue fusion, instruction-driven extraction, realistic evaluation, and trustworthy deployment.