CLIP-VAD: Exploiting Vision-Language Models for Voice Activity Detection

CLIPVLMRe-assign community