Infinite Talk + Index TTS - Master

Infinite Talk + Index TTS - Master is for audio, image-to-video, and video-to-video work. It takes audio and image inputs and includes Index TTS text and emotion controls, audio saving, and video combining.

The 33-node graph includes model loading, text encoding, sampling, decoding, image-to-video multi-talk, and video combining.

Index TTS nodes support base, emotion-audio, emotion-text, and emotion-vector controls.

Use LoadAudio and LoadImage for the listed audio and image inputs.

SaveAudio and VHS_VideoCombine provide the listed paths for saving audio and combining video.

The listed base model is Wan Video 14B i2v 480p.

Publisher: Sinusplusminus. Source page: https://civitai.com/models/2202910. No model-use license is listed.

Listed node packs: ComfyUI_Comfyroll_CustomNodes, comfyui-impact-pack, comfyui-resolution-master, comfyui-workflow-encrypt, and was-ns.

Listed model files: clip_vision_h.safetensors, lightx2v_I2V_14B_480p_cfg_step_distill_rank128_bf16.safetensors, model.safetensors, and umt5-xxl-enc-bf16.safetensors.

Additional listed model files: wan_2.1_vae.safetensors, Wan2_1-InfiniteTalk_Single_Q8.gguf, and wan2.1-i2v-14b-480p-Q4_K_M.gguf.

ComfyUI_Comfyroll_CustomNodes is pinned to d78b780ae43fcf8c6b7c6505e6ffb4584281ceca, and comfyui-impact-pack is pinned to e22f68fb97115b24ddf35d3d8801cc99fdb8af8d.

comfyui-resolution-master is pinned to 1.4.7, comfyui-workflow-encrypt to 1.0.0, and was-ns is listed at 3.0.1.

Before use, verify that model.safetensors, wan2.1-i2v-14b-480p-Q4_K_M.gguf, clip_vision_h.safetensors, and wan_2.1_vae.safetensors are the intended files.

Suggestion · not verified

Before use, verify that Wan2_1-InfiniteTalk_Single_Q8.gguf and umt5-xxl-enc-bf16.safetensors are the intended files, and that was-ns is the intended node pack.

Suggestion · not verified