Shifting Focus to Global Audio Perception in Portrait Animation

Sonic ComfyUI: https://github.com/smthemex/ComfyUI_Sonic

ComfyUI-EdgeTTS: https://github.com/1038lab/ComfyUI-EdgeTTS

ComfyUI-KokoroTTS: https://github.com/1038lab/ComfyUI-KokoroTTS

This model demands a substantial amount of VRAM. When utilizing an RTX4090 graphics card, generating a video typically takes approximately 16 minutes and 14 seconds to generate a 13 second video . The workflow is straightforward: simply input a single image and employ EdgeTTS to create an audio file. For those interested in implementing this workflow, it is essential to ensure that your system has at least 12GB of VRAM. Although the output quality is great, but it's slow.

If this node is helpful to you or if you like my work, please give a ❤️! It’s the greatest encouragement for my efforts!