Upload 11 files

.gitattributes

@@ -46,3 +46,7 @@ Workflow/silent_audio/silent_3sec.wav filter=lfs diff=lfs merge=lfs -text
 Workflow/silent_audio/silent_4sec.wav filter=lfs diff=lfs merge=lfs -text
 Workflow/silent_audio/silent_5sec.wav filter=lfs diff=lfs merge=lfs -text
 video/ComfyUI__00003-audio.mp4 filter=lfs diff=lfs merge=lfs -text
+ComfyUI-WanBerniniS2V_v2/demo_assets/ComfyUI_00207-audio.mp4 filter=lfs diff=lfs merge=lfs -text
+ComfyUI-WanBerniniS2V_v2/demo_assets/dude.jpg filter=lfs diff=lfs merge=lfs -text
+ComfyUI-WanBerniniS2V_v2/demo_assets/I[[:space:]]am[[:space:]]The[[:space:]]Dude[[:space:]]Playing[[:space:]]The[[:space:]]Dude.wav filter=lfs diff=lfs merge=lfs -text
+ComfyUI-WanBerniniS2V_v2/demo_assets/Im[[:space:]]the[[:space:]]Dude.wav filter=lfs diff=lfs merge=lfs -text

ComfyUI-WanBerniniS2V_v2/ComfyUI-WanBerniniS2V_v2.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:790d32a15fe19a572ab199ff9c9f7ae0c3099ba3169f3c9a51538a723700fa84
+size 2524040

ComfyUI-WanBerniniS2V_v2/Workflow/Bernini-R-S2V-Workflow-V2.json

@@ -0,0 +1,2857 @@
+{
+  "id": "ea7ba43e-ab56-40d4-a298-423f5bfe4462",
+  "revision": 0,
+  "last_node_id": 279,
+  "last_link_id": 758,
+  "nodes": [
+    {
+      "id": 80,
+      "type": "VAEDecode",
+      "pos": [
+        1210,
+        170
+      ],
+      "size": [
+        210,
+        50
+      ],
+      "flags": {},
+      "order": 41,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "samples",
+          "type": "LATENT",
+          "link": 533
+        },
+        {
+          "name": "vae",
+          "type": "VAE",
+          "link": 285
+        }
+      ],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            580
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "VAEDecode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 38,
+      "type": "CLIPLoader",
+      "pos": [
+        -530,
+        470
+      ],
+      "size": [
+        390,
+        110
+      ],
+      "flags": {},
+      "order": 0,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "CLIP",
+          "type": "CLIP",
+          "slot_index": 0,
+          "links": [
+            74,
+            75
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "CLIPLoader",
+        "models": [
+          {
+            "name": "umt5_xxl_fp8_e4m3fn_scaled.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors",
+            "directory": "text_encoders"
+          }
+        ]
+      },
+      "widgets_values": [
+        "umt5_xxl_fp8_e4m3fn_scaled.safetensors",
+        "wan",
+        "default"
+      ]
+    },
+    {
+      "id": 201,
+      "type": "RIFEInterpolation",
+      "pos": [
+        1480,
+        170
+      ],
+      "size": [
+        270,
+        180
+      ],
+      "flags": {},
+      "order": 42,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "images",
+          "type": "IMAGE",
+          "link": 580
+        }
+      ],
+      "outputs": [
+        {
+          "name": "images",
+          "type": "IMAGE",
+          "links": [
+            629
+          ]
+        }
+      ],
+      "properties": {
+        "aux_id": "GACLove/ComfyUI-VFI",
+        "ver": "6176a430f12cd16003f4664c1e3c6af8e96cc3c6",
+        "Node name for S&R": "RIFEInterpolation"
+      },
+      "widgets_values": [
+        16,
+        24,
+        2,
+        "flownet.pkl",
+        4,
+        true
+      ]
+    },
+    {
+      "id": 210,
+      "type": "LoraLoaderModelOnly",
+      "pos": [
+        1560,
+        450
+      ],
+      "size": [
+        360,
+        90
+      ],
+      "flags": {},
+      "order": 18,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 586
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "links": [
+            581
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "LoraLoaderModelOnly",
+        "models": [
+          {
+            "name": "wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/loras/wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors",
+            "directory": "loras"
+          }
+        ]
+      },
+      "widgets_values": [
+        "Wan-Speed/Bernini-R/Bernini-R_LightX2V_low_noise.safetensors",
+        1
+      ]
+    },
+    {
+      "id": 54,
+      "type": "ModelSamplingSD3",
+      "pos": [
+        1140,
+        590
+      ],
+      "size": [
+        370,
+        60
+      ],
+      "flags": {},
+      "order": 25,
+      "mode": 4,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 363
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            522
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "ModelSamplingSD3"
+      },
+      "widgets_values": [
+        8
+      ]
+    },
+    {
+      "id": 214,
+      "type": "ModelSamplingSD3",
+      "pos": [
+        1560,
+        590
+      ],
+      "size": [
+        360,
+        60
+      ],
+      "flags": {},
+      "order": 26,
+      "mode": 4,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 581
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            582
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "ModelSamplingSD3"
+      },
+      "widgets_values": [
+        8
+      ]
+    },
+    {
+      "id": 207,
+      "type": "KSamplerAdvanced",
+      "pos": [
+        1560,
+        690
+      ],
+      "size": [
+        360,
+        340
+      ],
+      "flags": {},
+      "order": 40,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 582
+        },
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 741
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 743
+        },
+        {
+          "name": "latent_image",
+          "type": "LATENT",
+          "link": 526
+        }
+      ],
+      "outputs": [
+        {
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            533
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "KSamplerAdvanced"
+      },
+      "widgets_values": [
+        "disable",
+        0,
+        "fixed",
+        4,
+        1,
+        "dpmpp_2m_sde",
+        "sgm_uniform",
+        2,
+        4,
+        "disable"
+      ]
+    },
+    {
+      "id": 250,
+      "type": "StringConcatenate",
+      "pos": [
+        -520,
+        930
+      ],
+      "size": [
+        400,
+        250
+      ],
+      "flags": {},
+      "order": 24,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "string_a",
+          "type": "STRING",
+          "widget": {
+            "name": "string_a"
+          },
+          "link": 662
+        },
+        {
+          "name": "string_b",
+          "type": "STRING",
+          "widget": {
+            "name": "string_b"
+          },
+          "link": 664
+        }
+      ],
+      "outputs": [
+        {
+          "name": "STRING",
+          "type": "STRING",
+          "links": [
+            663
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.24.0",
+        "Node name for S&R": "StringConcatenate"
+      },
+      "widgets_values": [
+        "",
+        "",
+        "."
+      ]
+    },
+    {
+      "id": 6,
+      "type": "CLIPTextEncode",
+      "pos": [
+        -50,
+        750
+      ],
+      "size": [
+        670,
+        280
+      ],
+      "flags": {
+        "collapsed": true
+      },
+      "order": 30,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 74
+        },
+        {
+          "name": "text",
+          "type": "STRING",
+          "widget": {
+            "name": "text"
+          },
+          "link": 663
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "slot_index": 0,
+          "links": [
+            725
+          ]
+        }
+      ],
+      "title": "CLIP Text Encode (Positive Prompt)",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "CLIPTextEncode"
+      },
+      "widgets_values": [
+        ""
+      ],
+      "color": "#232",
+      "bgcolor": "#353"
+    },
+    {
+      "id": 251,
+      "type": "CustomCombo",
+      "pos": [
+        -1020,
+        760
+      ],
+      "size": [
+        460,
+        390
+      ],
+      "flags": {},
+      "order": 1,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "STRING",
+          "type": "STRING",
+          "links": []
+        },
+        {
+          "name": "INDEX",
+          "type": "INT",
+          "links": [
+            661
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.24.0",
+        "Node name for S&R": "CustomCombo"
+      },
+      "widgets_values": [
+        "Image to Video",
+        5,
+        "Default",
+        "Text to Image",
+        "Text to Video",
+        "Image Editing",
+        "Subject to Image",
+        "Image to Video",
+        "Video Editing",
+        "Video Editing (Content Propagation)",
+        "Video Editing with Reference",
+        "Ads / Content Insertion",
+        "Video Editing (Action / Position)",
+        "Video Editing (Style / Motion)",
+        ""
+      ]
+    },
+    {
+      "id": 249,
+      "type": "6da2792c-a953-4960-a53a-ff1b5df6fe09",
+      "pos": [
+        -530,
+        750
+      ],
+      "size": [
+        410,
+        150
+      ],
+      "flags": {},
+      "order": 15,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "index",
+          "type": "INT",
+          "widget": {
+            "name": "index"
+          },
+          "link": 661
+        }
+      ],
+      "outputs": [
+        {
+          "name": "selected_line",
+          "type": "STRING",
+          "links": [
+            662
+          ]
+        }
+      ],
+      "properties": {
+        "proxyWidgets": [
+          [
+            "2",
+            "string"
+          ],
+          [
+            "248",
+            "value"
+          ]
+        ],
+        "cnr_id": "comfy-core",
+        "ver": "0.19.0",
+        "ue_properties": {
+          "widget_ue_connectable": {},
+          "input_ue_unconnectable": {}
+        }
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 107,
+      "type": "LoraLoaderModelOnly",
+      "pos": [
+        1140,
+        450
+      ],
+      "size": [
+        380,
+        90
+      ],
+      "flags": {},
+      "order": 17,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 585
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "links": [
+            363
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "LoraLoaderModelOnly",
+        "models": [
+          {
+            "name": "wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/loras/wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors",
+            "directory": "loras"
+          }
+        ]
+      },
+      "widgets_values": [
+        "Wan-Speed/Bernini-R/Bernini-R_LightX2V_high_noise.safetensors",
+        1
+      ]
+    },
+    {
+      "id": 205,
+      "type": "KSamplerAdvanced",
+      "pos": [
+        1130,
+        710
+      ],
+      "size": [
+        370,
+        340
+      ],
+      "flags": {},
+      "order": 38,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 522
+        },
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 740
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 742
+        },
+        {
+          "name": "latent_image",
+          "type": "LATENT",
+          "link": 744
+        }
+      ],
+      "outputs": [
+        {
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            526
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "KSamplerAdvanced"
+      },
+      "widgets_values": [
+        "enable",
+        780018084687208,
+        "randomize",
+        4,
+        1,
+        "dpmpp_2m_sde",
+        "sgm_uniform",
+        0,
+        2,
+        "enable"
+      ]
+    },
+    {
+      "id": 271,
+      "type": "Reroute",
+      "pos": [
+        -360,
+        1610
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 29,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 737
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO",
+          "links": [
+            738,
+            739
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      },
+      "color": "#323",
+      "bgcolor": "#535"
+    },
+    {
+      "id": 39,
+      "type": "VAELoader",
+      "pos": [
+        530,
+        730
+      ],
+      "size": [
+        210,
+        60
+      ],
+      "flags": {
+        "collapsed": true
+      },
+      "order": 2,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "VAE",
+          "type": "VAE",
+          "slot_index": 0,
+          "links": [
+            285,
+            727
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "VAELoader",
+        "models": [
+          {
+            "name": "wan_2.1_vae.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/vae/wan_2.1_vae.safetensors",
+            "directory": "vae"
+          }
+        ]
+      },
+      "widgets_values": [
+        "Wan/Wan2_1_VAE_fp32.safetensors"
+      ]
+    },
+    {
+      "id": 235,
+      "type": "Reroute",
+      "pos": [
+        -350,
+        1410
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 28,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 736
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO",
+          "links": [
+            622,
+            717
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      },
+      "color": "#323",
+      "bgcolor": "#535"
+    },
+    {
+      "id": 270,
+      "type": "Reroute",
+      "pos": [
+        -450,
+        1290
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 19,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 733
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO_ENCODER",
+          "links": [
+            734,
+            735
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      }
+    },
+    {
+      "id": 241,
+      "type": "LoadImage",
+      "pos": [
+        -40,
+        840
+      ],
+      "size": [
+        290,
+        320
+      ],
+      "flags": {},
+      "order": 3,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            731
+          ]
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": [
+            702,
+            729
+          ]
+        }
+      ],
+      "title": "image0 & mask_1",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "LoadImage",
+        "image": "clipspace/clipspace-painted-masked-1783665396446.png [input]"
+      },
+      "widgets_values": [
+        "clipspace/clipspace-painted-masked-1783665396446.png [input]",
+        "image"
+      ]
+    },
+    {
+      "id": 265,
+      "type": "InvertMask",
+      "pos": [
+        240,
+        790
+      ],
+      "size": [
+        140,
+        30
+      ],
+      "flags": {
+        "collapsed": true
+      },
+      "order": 16,
+      "mode": 4,
+      "inputs": [
+        {
+          "name": "mask",
+          "type": "MASK",
+          "link": 702
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": []
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "InvertMask"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 7,
+      "type": "CLIPTextEncode",
+      "pos": [
+        10,
+        500
+      ],
+      "size": [
+        650,
+        180
+      ],
+      "flags": {
+        "collapsed": true
+      },
+      "order": 14,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 75
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "slot_index": 0,
+          "links": [
+            726
+          ]
+        }
+      ],
+      "title": "CLIP Text Encode (Negative Prompt)",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "CLIPTextEncode"
+      },
+      "widgets_values": [
+        "Vivid color tone, overexposed, static, unclear details, subtitles, style, artwork, painting, image, still, overall grayish, worst quality, low quality, leftover JPEG compression artifacts, ugly, incomplete, missing parts, extra fingers, poorly drawn hands, poorly drawn face, disfigured, malformed body parts, fused fingers, a completely motionless image, messy background, three legs, many people in the background, walking backward."
+      ],
+      "color": "#223",
+      "bgcolor": "#335"
+    },
+    {
+      "id": 37,
+      "type": "UNETLoader",
+      "pos": [
+        -530,
+        180
+      ],
+      "size": [
+        390,
+        90
+      ],
+      "flags": {},
+      "order": 4,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            585
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "UNETLoader",
+        "models": [
+          {
+            "name": "wan2.2_s2v_14B_fp8_scaled.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/diffusion_models/wan2.2_s2v_14B_fp8_scaled.safetensors",
+            "directory": "diffusion_models"
+          }
+        ]
+      },
+      "widgets_values": [
+        "Bernini-R_fp8/Bernini-R-S2V-FP8/wan2.2_bernini_r_high_noise_fp8_scaled_s2v.safetensors",
+        "default"
+      ]
+    },
+    {
+      "id": 204,
+      "type": "UNETLoader",
+      "pos": [
+        -530,
+        320
+      ],
+      "size": [
+        390,
+        90
+      ],
+      "flags": {},
+      "order": 5,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "slot_index": 0,
+          "links": [
+            586
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "UNETLoader",
+        "models": [
+          {
+            "name": "wan2.2_s2v_14B_fp8_scaled.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/diffusion_models/wan2.2_s2v_14B_fp8_scaled.safetensors",
+            "directory": "diffusion_models"
+          }
+        ]
+      },
+      "widgets_values": [
+        "Bernini-R_fp8/Bernini-R-S2V-FP8/wan2.2_bernini_r_low_noise_fp8_scaled_s2v.safetensors",
+        "default"
+      ]
+    },
+    {
+      "id": 267,
+      "type": "AudioConcat",
+      "pos": [
+        -210,
+        1290
+      ],
+      "size": [
+        290,
+        80
+      ],
+      "flags": {},
+      "order": 34,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio1",
+          "type": "AUDIO",
+          "link": 717
+        },
+        {
+          "name": "audio2",
+          "type": "AUDIO",
+          "link": 739
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            745,
+            754
+          ]
+        }
+      ],
+      "title": "audio_1 + audio_2",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "AudioConcat"
+      },
+      "widgets_values": [
+        "after"
+      ]
+    },
+    {
+      "id": 238,
+      "type": "ComfyMathExpression",
+      "pos": [
+        1110,
+        1110
+      ],
+      "size": [
+        400,
+        200
+      ],
+      "flags": {
+        "collapsed": true
+      },
+      "order": 21,
+      "mode": 0,
+      "inputs": [
+        {
+          "label": "a",
+          "name": "values.a",
+          "type": "FLOAT,INT,BOOLEAN",
+          "link": 627
+        },
+        {
+          "label": "b",
+          "name": "values.b",
+          "shape": 7,
+          "type": "FLOAT,INT,BOOLEAN",
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "name": "FLOAT",
+          "type": "FLOAT",
+          "links": null
+        },
+        {
+          "name": "INT",
+          "type": "INT",
+          "links": [
+            641,
+            724
+          ]
+        },
+        {
+          "name": "BOOL",
+          "type": "BOOLEAN",
+          "links": null
+        }
+      ],
+      "title": "16 F/S",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "ComfyMathExpression"
+      },
+      "widgets_values": [
+        "a * 16 + 1"
+      ]
+    },
+    {
+      "id": 57,
+      "type": "AudioEncoderLoader",
+      "pos": [
+        -1030,
+        1290
+      ],
+      "size": [
+        290,
+        60
+      ],
+      "flags": {},
+      "order": 6,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "AUDIO_ENCODER",
+          "type": "AUDIO_ENCODER",
+          "links": [
+            733
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "AudioEncoderLoader",
+        "models": [
+          {
+            "name": "wav2vec2_large_english_fp16.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/audio_encoders/wav2vec2_large_english_fp16.safetensors",
+            "directory": "audio_encoders"
+          }
+        ]
+      },
+      "widgets_values": [
+        "wav2vec2_large_english_fp16.safetensors"
+      ]
+    },
+    {
+      "id": 266,
+      "type": "AudioEncoderEncode",
+      "pos": [
+        -220,
+        1610
+      ],
+      "size": [
+        290,
+        50
+      ],
+      "flags": {},
+      "order": 33,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio_encoder",
+          "type": "AUDIO_ENCODER",
+          "link": 735
+        },
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 738
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO_ENCODER_OUTPUT",
+          "type": "AUDIO_ENCODER_OUTPUT",
+          "links": [
+            730
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "AudioEncoderEncode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 275,
+      "type": "AudioEncoderLoader",
+      "pos": [
+        120,
+        1310
+      ],
+      "size": [
+        290,
+        60
+      ],
+      "flags": {},
+      "order": 7,
+      "mode": 2,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "AUDIO_ENCODER",
+          "type": "AUDIO_ENCODER",
+          "links": [
+            746
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "AudioEncoderLoader",
+        "models": [
+          {
+            "name": "wav2vec2_large_english_fp16.safetensors",
+            "url": "https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/audio_encoders/wav2vec2_large_english_fp16.safetensors",
+            "directory": "audio_encoders"
+          }
+        ]
+      },
+      "widgets_values": [
+        "wav2vec2_large_english_fp16.safetensors"
+      ]
+    },
+    {
+      "id": 273,
+      "type": "LoadAudio",
+      "pos": [
+        120,
+        1420
+      ],
+      "size": [
+        290,
+        140
+      ],
+      "flags": {},
+      "order": 8,
+      "mode": 2,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            747
+          ]
+        }
+      ],
+      "title": "audio_1",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "LoadAudio"
+      },
+      "widgets_values": [
+        "anthony.wav",
+        null,
+        null
+      ]
+    },
+    {
+      "id": 274,
+      "type": "TrimAudioDuration",
+      "pos": [
+        440,
+        1420
+      ],
+      "size": [
+        270,
+        90
+      ],
+      "flags": {},
+      "order": 20,
+      "mode": 2,
+      "inputs": [
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 747
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            751
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "TrimAudioDuration"
+      },
+      "widgets_values": [
+        0,
+        9
+      ]
+    },
+    {
+      "id": 276,
+      "type": "AudioEncoderEncode",
+      "pos": [
+        740,
+        1300
+      ],
+      "size": [
+        290,
+        50
+      ],
+      "flags": {},
+      "order": 31,
+      "mode": 2,
+      "inputs": [
+        {
+          "name": "audio_encoder",
+          "type": "AUDIO_ENCODER",
+          "link": 746
+        },
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 752
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO_ENCODER_OUTPUT",
+          "type": "AUDIO_ENCODER_OUTPUT",
+          "links": []
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "AudioEncoderEncode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 279,
+      "type": "Reroute",
+      "pos": [
+        1920,
+        250
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 39,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 757
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO",
+          "links": [
+            755
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      },
+      "color": "#323",
+      "bgcolor": "#535"
+    },
+    {
+      "id": 277,
+      "type": "Reroute",
+      "pos": [
+        770,
+        1430
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 27,
+      "mode": 2,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 751
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO",
+          "links": [
+            752
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      },
+      "color": "#323",
+      "bgcolor": "#535"
+    },
+    {
+      "id": 239,
+      "type": "PrimitiveInt",
+      "pos": [
+        710,
+        1080
+      ],
+      "size": [
+        350,
+        90
+      ],
+      "flags": {},
+      "order": 9,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "INT",
+          "type": "INT",
+          "links": [
+            627
+          ]
+        }
+      ],
+      "title": "Video Length in Seconds",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "PrimitiveInt"
+      },
+      "widgets_values": [
+        9,
+        "fixed"
+      ]
+    },
+    {
+      "id": 269,
+      "type": "BerniniS2VConditioningV2",
+      "pos": [
+        710,
+        600
+      ],
+      "size": [
+        360,
+        450
+      ],
+      "flags": {},
+      "order": 35,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 725
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 726
+        },
+        {
+          "name": "vae",
+          "type": "VAE",
+          "link": 727
+        },
+        {
+          "name": "audio_1",
+          "type": "AUDIO_ENCODER_OUTPUT",
+          "link": 758
+        },
+        {
+          "name": "mask_1",
+          "type": "MASK",
+          "link": 729
+        },
+        {
+          "name": "audio_2",
+          "shape": 7,
+          "type": "AUDIO_ENCODER_OUTPUT",
+          "link": 730
+        },
+        {
+          "name": "mask_2",
+          "shape": 7,
+          "type": "MASK",
+          "link": 732
+        },
+        {
+          "name": "source_video",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": null
+        },
+        {
+          "name": "reference_video",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": null
+        },
+        {
+          "label": "reference_image_0",
+          "name": "reference_images.reference_image_0",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": 731
+        },
+        {
+          "label": "reference_image_1",
+          "name": "reference_images.reference_image_1",
+          "shape": 7,
+          "type": "IMAGE",
+          "link": null
+        },
+        {
+          "name": "length",
+          "type": "INT",
+          "widget": {
+            "name": "length"
+          },
+          "link": 724
+        }
+      ],
+      "outputs": [
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "links": [
+            740,
+            741
+          ]
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "links": [
+            742,
+            743
+          ]
+        },
+        {
+          "name": "latent",
+          "type": "LATENT",
+          "links": [
+            744
+          ]
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "BerniniS2VConditioningV2"
+      },
+      "widgets_values": [
+        720,
+        480,
+        81,
+        1,
+        -1,
+        0,
+        1,
+        848
+      ]
+    },
+    {
+      "id": 240,
+      "type": "VHS_VideoCombine",
+      "pos": [
+        2040,
+        230
+      ],
+      "size": [
+        470,
+        648
+      ],
+      "flags": {},
+      "order": 43,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "images",
+          "type": "IMAGE",
+          "link": 629
+        },
+        {
+          "name": "audio",
+          "shape": 7,
+          "type": "AUDIO",
+          "link": 755
+        },
+        {
+          "name": "meta_batch",
+          "shape": 7,
+          "type": "VHS_BatchManager",
+          "link": null
+        },
+        {
+          "name": "vae",
+          "shape": 7,
+          "type": "VAE",
+          "link": null
+        }
+      ],
+      "outputs": [
+        {
+          "name": "Filenames",
+          "type": "VHS_FILENAMES",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfyui-videohelpersuite",
+        "ver": "1.7.9",
+        "Node name for S&R": "VHS_VideoCombine"
+      },
+      "widgets_values": {
+        "frame_rate": 24,
+        "loop_count": 0,
+        "filename_prefix": "video/ComfyUI",
+        "format": "video/h264-mp4",
+        "pix_fmt": "yuv420p",
+        "crf": 19,
+        "save_metadata": true,
+        "trim_to_audio": true,
+        "pingpong": false,
+        "save_output": true,
+        "videopreview": {
+          "hidden": false,
+          "paused": false,
+          "params": {
+            "filename": "ComfyUI_00207-audio.mp4",
+            "subfolder": "video",
+            "type": "output",
+            "format": "video/h264-mp4",
+            "frame_rate": 24,
+            "workflow": "ComfyUI_00207.png",
+            "fullpath": "/media/dgt/N33B/AI/ComfyUI/output/video/ComfyUI_00207-audio.mp4"
+          }
+        }
+      }
+    },
+    {
+      "id": 268,
+      "type": "LoadImage",
+      "pos": [
+        320,
+        840
+      ],
+      "size": [
+        290,
+        320
+      ],
+      "flags": {},
+      "order": 10,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": []
+        },
+        {
+          "name": "MASK",
+          "type": "MASK",
+          "links": [
+            732
+          ]
+        }
+      ],
+      "title": "Mask_2",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "LoadImage",
+        "image": "clipspace/clipspace-painted-masked-1783665412223.png [input]"
+      },
+      "widgets_values": [
+        "clipspace/clipspace-painted-masked-1783665412223.png [input]",
+        "image"
+      ]
+    },
+    {
+      "id": 278,
+      "type": "Reroute",
+      "pos": [
+        0,
+        1220
+      ],
+      "size": [
+        75,
+        26
+      ],
+      "flags": {},
+      "order": 37,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "",
+          "type": "*",
+          "link": 754
+        }
+      ],
+      "outputs": [
+        {
+          "name": "",
+          "type": "AUDIO",
+          "links": [
+            757
+          ]
+        }
+      ],
+      "properties": {
+        "showOutputText": false,
+        "horizontal": false
+      },
+      "color": "#323",
+      "bgcolor": "#535"
+    },
+    {
+      "id": 58,
+      "type": "LoadAudio",
+      "pos": [
+        -1030,
+        1410
+      ],
+      "size": [
+        290,
+        140
+      ],
+      "flags": {},
+      "order": 11,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            671
+          ]
+        }
+      ],
+      "title": "audio_1",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "LoadAudio"
+      },
+      "widgets_values": [
+        "I am The Dude Playing The Dude.wav",
+        null,
+        null
+      ]
+    },
+    {
+      "id": 253,
+      "type": "TrimAudioDuration",
+      "pos": [
+        -660,
+        1610
+      ],
+      "size": [
+        270,
+        90
+      ],
+      "flags": {},
+      "order": 23,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 667
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            737
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "TrimAudioDuration"
+      },
+      "widgets_values": [
+        0,
+        3
+      ]
+    },
+    {
+      "id": 223,
+      "type": "TrimAudioDuration",
+      "pos": [
+        -660,
+        1410
+      ],
+      "size": [
+        270,
+        90
+      ],
+      "flags": {},
+      "order": 22,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 671
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            736
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "TrimAudioDuration"
+      },
+      "widgets_values": [
+        0,
+        6
+      ]
+    },
+    {
+      "id": 247,
+      "type": "LoadAudio",
+      "pos": [
+        -1030,
+        1610
+      ],
+      "size": [
+        290,
+        140
+      ],
+      "flags": {},
+      "order": 12,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "AUDIO",
+          "type": "AUDIO",
+          "links": [
+            667
+          ]
+        }
+      ],
+      "title": "audio_2",
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "LoadAudio"
+      },
+      "widgets_values": [
+        "Im the Dude.wav",
+        null,
+        null
+      ]
+    },
+    {
+      "id": 252,
+      "type": "PrimitiveStringMultiline",
+      "pos": [
+        -10,
+        170
+      ],
+      "size": [
+        710,
+        290
+      ],
+      "flags": {},
+      "order": 13,
+      "mode": 0,
+      "inputs": [],
+      "outputs": [
+        {
+          "name": "STRING",
+          "type": "STRING",
+          "links": [
+            664
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "PrimitiveStringMultiline"
+      },
+      "widgets_values": [
+        "two men in image0 sitting in a restaurant."
+      ],
+      "color": "#232",
+      "bgcolor": "#353"
+    },
+    {
+      "id": 56,
+      "type": "AudioEncoderEncode",
+      "pos": [
+        -210,
+        1410
+      ],
+      "size": [
+        290,
+        50
+      ],
+      "flags": {},
+      "order": 32,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio_encoder",
+          "type": "AUDIO_ENCODER",
+          "link": 734
+        },
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 622
+        }
+      ],
+      "outputs": [
+        {
+          "name": "AUDIO_ENCODER_OUTPUT",
+          "type": "AUDIO_ENCODER_OUTPUT",
+          "links": [
+            758
+          ]
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.3.54",
+        "Node name for S&R": "AudioEncoderEncode"
+      },
+      "widgets_values": []
+    },
+    {
+      "id": 272,
+      "type": "PreviewAudio",
+      "pos": [
+        -180,
+        1720
+      ],
+      "size": [
+        270,
+        90
+      ],
+      "flags": {},
+      "order": 36,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "link": 745
+        }
+      ],
+      "outputs": [
+        {
+          "name": "audio",
+          "type": "AUDIO",
+          "links": null
+        }
+      ],
+      "properties": {
+        "cnr_id": "comfy-core",
+        "ver": "0.27.0",
+        "Node name for S&R": "PreviewAudio"
+      },
+      "widgets_values": []
+    }
+  ],
+  "links": [
+    [
+      74,
+      38,
+      0,
+      6,
+      0,
+      "CLIP"
+    ],
+    [
+      75,
+      38,
+      0,
+      7,
+      0,
+      "CLIP"
+    ],
+    [
+      285,
+      39,
+      0,
+      80,
+      1,
+      "VAE"
+    ],
+    [
+      363,
+      107,
+      0,
+      54,
+      0,
+      "MODEL"
+    ],
+    [
+      522,
+      54,
+      0,
+      205,
+      0,
+      "MODEL"
+    ],
+    [
+      526,
+      205,
+      0,
+      207,
+      3,
+      "LATENT"
+    ],
+    [
+      533,
+      207,
+      0,
+      80,
+      0,
+      "LATENT"
+    ],
+    [
+      580,
+      80,
+      0,
+      201,
+      0,
+      "IMAGE"
+    ],
+    [
+      581,
+      210,
+      0,
+      214,
+      0,
+      "MODEL"
+    ],
+    [
+      582,
+      214,
+      0,
+      207,
+      0,
+      "MODEL"
+    ],
+    [
+      585,
+      37,
+      0,
+      107,
+      0,
+      "MODEL"
+    ],
+    [
+      586,
+      204,
+      0,
+      210,
+      0,
+      "MODEL"
+    ],
+    [
+      622,
+      235,
+      0,
+      56,
+      1,
+      "AUDIO"
+    ],
+    [
+      627,
+      239,
+      0,
+      238,
+      0,
+      "INT"
+    ],
+    [
+      629,
+      201,
+      0,
+      240,
+      0,
+      "IMAGE"
+    ],
+    [
+      661,
+      251,
+      1,
+      249,
+      0,
+      "INT"
+    ],
+    [
+      662,
+      249,
+      0,
+      250,
+      0,
+      "STRING"
+    ],
+    [
+      663,
+      250,
+      0,
+      6,
+      1,
+      "STRING"
+    ],
+    [
+      664,
+      252,
+      0,
+      250,
+      1,
+      "STRING"
+    ],
+    [
+      667,
+      247,
+      0,
+      253,
+      0,
+      "AUDIO"
+    ],
+    [
+      671,
+      58,
+      0,
+      223,
+      0,
+      "AUDIO"
+    ],
+    [
+      702,
+      241,
+      1,
+      265,
+      0,
+      "MASK"
+    ],
+    [
+      717,
+      235,
+      0,
+      267,
+      0,
+      "AUDIO"
+    ],
+    [
+      724,
+      238,
+      1,
+      269,
+      11,
+      "INT"
+    ],
+    [
+      725,
+      6,
+      0,
+      269,
+      0,
+      "CONDITIONING"
+    ],
+    [
+      726,
+      7,
+      0,
+      269,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      727,
+      39,
+      0,
+      269,
+      2,
+      "VAE"
+    ],
+    [
+      729,
+      241,
+      1,
+      269,
+      4,
+      "MASK"
+    ],
+    [
+      730,
+      266,
+      0,
+      269,
+      5,
+      "AUDIO_ENCODER_OUTPUT"
+    ],
+    [
+      731,
+      241,
+      0,
+      269,
+      9,
+      "IMAGE"
+    ],
+    [
+      732,
+      268,
+      1,
+      269,
+      6,
+      "MASK"
+    ],
+    [
+      733,
+      57,
+      0,
+      270,
+      0,
+      "AUDIO_ENCODER"
+    ],
+    [
+      734,
+      270,
+      0,
+      56,
+      0,
+      "AUDIO_ENCODER"
+    ],
+    [
+      735,
+      270,
+      0,
+      266,
+      0,
+      "AUDIO_ENCODER"
+    ],
+    [
+      736,
+      223,
+      0,
+      235,
+      0,
+      "AUDIO"
+    ],
+    [
+      737,
+      253,
+      0,
+      271,
+      0,
+      "AUDIO"
+    ],
+    [
+      738,
+      271,
+      0,
+      266,
+      1,
+      "AUDIO"
+    ],
+    [
+      739,
+      271,
+      0,
+      267,
+      1,
+      "AUDIO"
+    ],
+    [
+      740,
+      269,
+      0,
+      205,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      741,
+      269,
+      0,
+      207,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      742,
+      269,
+      1,
+      205,
+      2,
+      "CONDITIONING"
+    ],
+    [
+      743,
+      269,
+      1,
+      207,
+      2,
+      "CONDITIONING"
+    ],
+    [
+      744,
+      269,
+      2,
+      205,
+      3,
+      "LATENT"
+    ],
+    [
+      745,
+      267,
+      0,
+      272,
+      0,
+      "AUDIO"
+    ],
+    [
+      746,
+      275,
+      0,
+      276,
+      0,
+      "AUDIO_ENCODER"
+    ],
+    [
+      747,
+      273,
+      0,
+      274,
+      0,
+      "AUDIO"
+    ],
+    [
+      751,
+      274,
+      0,
+      277,
+      0,
+      "AUDIO"
+    ],
+    [
+      752,
+      277,
+      0,
+      276,
+      1,
+      "AUDIO"
+    ],
+    [
+      754,
+      267,
+      0,
+      278,
+      0,
+      "AUDIO"
+    ],
+    [
+      755,
+      279,
+      0,
+      240,
+      1,
+      "AUDIO"
+    ],
+    [
+      757,
+      278,
+      0,
+      279,
+      0,
+      "AUDIO"
+    ],
+    [
+      758,
+      56,
+      0,
+      269,
+      3,
+      "AUDIO_ENCODER_OUTPUT"
+    ]
+  ],
+  "groups": [
+    {
+      "id": 2,
+      "title": "-",
+      "bounding": [
+        -110,
+        100,
+        2650,
+        1090
+      ],
+      "color": "#3f789e",
+      "flags": {}
+    },
+    {
+      "id": 3,
+      "title": "-",
+      "bounding": [
+        -540,
+        100,
+        430,
+        490
+      ],
+      "color": "#3f789e",
+      "flags": {}
+    },
+    {
+      "id": 10,
+      "title": "Tasks",
+      "bounding": [
+        -1030,
+        680,
+        920,
+        510
+      ],
+      "color": "#3f789e",
+      "flags": {}
+    },
+    {
+      "id": 11,
+      "title": "2 Speakers",
+      "bounding": [
+        -1040,
+        1220,
+        1130,
+        540
+      ],
+      "color": "#3f789e",
+      "flags": {}
+    },
+    {
+      "id": 12,
+      "title": "1 Speaker",
+      "bounding": [
+        90,
+        1220,
+        1170,
+        530
+      ],
+      "color": "#3f789e",
+      "flags": {}
+    }
+  ],
+  "definitions": {
+    "subgraphs": [
+      {
+        "id": "6da2792c-a953-4960-a53a-ff1b5df6fe09",
+        "version": 1,
+        "state": {
+          "lastGroupId": 12,
+          "lastNodeId": 279,
+          "lastLinkId": 758,
+          "lastRerouteId": 0
+        },
+        "revision": 0,
+        "config": {},
+        "name": "Select Per-Line Text by Index",
+        "description": "Selects one line from multiline text by zero-based index for batch or list-driven prompt workflows.",
+        "inputNode": {
+          "id": -10,
+          "bounding": [
+            -990,
+            8595,
+            128,
+            88
+          ]
+        },
+        "outputNode": {
+          "id": -20,
+          "bounding": [
+            710,
+            8585,
+            128,
+            68
+          ]
+        },
+        "inputs": [
+          {
+            "id": "75417d82-a934-4ac9-b667-d8dcd5a3bfb3",
+            "name": "text_per_line",
+            "type": "STRING",
+            "linkIds": [
+              13
+            ],
+            "localized_name": "text_per_line",
+            "pos": [
+              -886,
+              8619
+            ]
+          },
+          {
+            "id": "46e69a73-1804-4ca6-9175-31445bf0be96",
+            "name": "index",
+            "type": "INT",
+            "linkIds": [
+              14
+            ],
+            "localized_name": "index",
+            "pos": [
+              -886,
+              8639
+            ]
+          }
+        ],
+        "outputs": [
+          {
+            "id": "e34e8ad1-84d2-4bd2-a460-eb7de6067c10",
+            "name": "selected_line",
+            "type": "STRING",
+            "linkIds": [
+              10
+            ],
+            "localized_name": "selected_line",
+            "pos": [
+              734,
+              8609
+            ]
+          }
+        ],
+        "widgets": [],
+        "nodes": [
+          {
+            "id": 1,
+            "type": "PreviewAny",
+            "pos": [
+              -500,
+              8400
+            ],
+            "size": [
+              230,
+              180
+            ],
+            "flags": {},
+            "order": 0,
+            "mode": 0,
+            "inputs": [
+              {
+                "localized_name": "source",
+                "name": "source",
+                "type": "*",
+                "link": 1
+              }
+            ],
+            "outputs": [
+              {
+                "localized_name": "STRING",
+                "name": "STRING",
+                "type": "STRING",
+                "links": [
+                  6
+                ]
+              }
+            ],
+            "properties": {
+              "cnr_id": "comfy-core",
+              "ver": "0.19.0",
+              "Node name for S&R": "PreviewAny",
+              "ue_properties": {
+                "widget_ue_connectable": {},
+                "input_ue_unconnectable": {}
+              }
+            },
+            "widgets_values": [
+              null,
+              null,
+              null
+            ]
+          },
+          {
+            "id": 2,
+            "type": "RegexExtract",
+            "pos": [
+              -240,
+              8740
+            ],
+            "size": [
+              470,
+              460
+            ],
+            "flags": {},
+            "order": 1,
+            "mode": 0,
+            "showAdvanced": false,
+            "inputs": [
+              {
+                "localized_name": "string",
+                "name": "string",
+                "type": "STRING",
+                "widget": {
+                  "name": "string"
+                },
+                "link": 13
+              },
+              {
+                "localized_name": "regex_pattern",
+                "name": "regex_pattern",
+                "type": "STRING",
+                "widget": {
+                  "name": "regex_pattern"
+                },
+                "link": 9
+              }
+            ],
+            "outputs": [
+              {
+                "localized_name": "STRING",
+                "name": "STRING",
+                "type": "STRING",
+                "links": [
+                  10
+                ]
+              }
+            ],
+            "properties": {
+              "cnr_id": "comfy-core",
+              "ver": "0.19.0",
+              "Node name for S&R": "RegexExtract",
+              "ue_properties": {
+                "widget_ue_connectable": {},
+                "input_ue_unconnectable": {}
+              }
+            },
+            "widgets_values": [
+              "You are a helpful assistant.\nYou are a helpful assistant specialized in text-to-image generation.\nYou are a helpful assistant specialized in text-to-video generation.\nYou are a helpful assistant specialized in image editing.\nYou are a helpful assistant specialized in subject-to-image generation.\nYou are a helpful assistant specialized in image-to-video generation.\nYou are a helpful assistant specialized in video editing.\nYou are a helpful assistant specialized in video editing on content propagation.\nYou are a helpful assistant specialized in video editing with reference.\nYou are a helpful assistant specialized in ads insertion.\nYou are a helpful assistant for editing. You may need to adjust the subject's action or position.\nYou are a helpful assistant for editing. You might need to adjust the video's style, lighting, colors, textures, and the subject's pose or action.",
+              "",
+              "First Group",
+              false,
+              false,
+              false,
+              1
+            ]
+          },
+          {
+            "id": 248,
+            "type": "PrimitiveInt",
+            "pos": [
+              -810,
+              8400
+            ],
+            "size": [
+              270,
+              110
+            ],
+            "flags": {},
+            "order": 3,
+            "mode": 0,
+            "inputs": [
+              {
+                "localized_name": "value",
+                "name": "value",
+                "type": "INT",
+                "widget": {
+                  "name": "value"
+                },
+                "link": 14
+              }
+            ],
+            "outputs": [
+              {
+                "localized_name": "INT",
+                "name": "INT",
+                "type": "INT",
+                "links": [
+                  1
+                ]
+              }
+            ],
+            "title": "Int (line index)",
+            "properties": {
+              "cnr_id": "comfy-core",
+              "ver": "0.19.0",
+              "Node name for S&R": "Int (line index)",
+              "ue_properties": {
+                "widget_ue_connectable": {},
+                "input_ue_unconnectable": {}
+              }
+            },
+            "widgets_values": [
+              0,
+              "fixed"
+            ]
+          },
+          {
+            "id": 8,
+            "type": "StringReplace",
+            "pos": [
+              -240,
+              8400
+            ],
+            "size": [
+              400,
+              280
+            ],
+            "flags": {},
+            "order": 2,
+            "mode": 0,
+            "inputs": [
+              {
+                "localized_name": "replace",
+                "name": "replace",
+                "type": "STRING",
+                "widget": {
+                  "name": "replace"
+                },
+                "link": 6
+              }
+            ],
+            "outputs": [
+              {
+                "localized_name": "STRING",
+                "name": "STRING",
+                "type": "STRING",
+                "links": [
+                  9
+                ]
+              }
+            ],
+            "properties": {
+              "cnr_id": "comfy-core",
+              "ver": "0.19.0",
+              "Node name for S&R": "StringReplace",
+              "ue_properties": {
+                "widget_ue_connectable": {},
+                "input_ue_unconnectable": {}
+              }
+            },
+            "widgets_values": [
+              "^(?:[^\\n]*\\n){index}([^\\n]*)(?:\\n|$)",
+              "index",
+              ""
+            ]
+          }
+        ],
+        "groups": [],
+        "links": [
+          {
+            "id": 1,
+            "origin_id": 248,
+            "origin_slot": 0,
+            "target_id": 1,
+            "target_slot": 0,
+            "type": "INT"
+          },
+          {
+            "id": 9,
+            "origin_id": 8,
+            "origin_slot": 0,
+            "target_id": 2,
+            "target_slot": 1,
+            "type": "STRING"
+          },
+          {
+            "id": 6,
+            "origin_id": 1,
+            "origin_slot": 0,
+            "target_id": 8,
+            "target_slot": 0,
+            "type": "STRING"
+          },
+          {
+            "id": 10,
+            "origin_id": 2,
+            "origin_slot": 0,
+            "target_id": -20,
+            "target_slot": 0,
+            "type": "STRING"
+          },
+          {
+            "id": 13,
+            "origin_id": -10,
+            "origin_slot": 0,
+            "target_id": 2,
+            "target_slot": 0,
+            "type": "STRING"
+          },
+          {
+            "id": 14,
+            "origin_id": -10,
+            "origin_slot": 1,
+            "target_id": 248,
+            "target_slot": 0,
+            "type": "INT"
+          }
+        ],
+        "extra": {
+          "ue_links": [],
+          "links_added_by_ue": []
+        }
+      }
+    ]
+  },
+  "config": {},
+  "extra": {
+    "ds": {
+      "scale": 0.6618166065230274,
+      "offset": [
+        567.2760482151011,
+        4.735915514621264
+      ]
+    },
+    "frontendVersion": "1.45.20",
+    "VHS_latentpreview": false,
+    "VHS_latentpreviewrate": 0,
+    "VHS_MetadataImage": true,
+    "VHS_KeepIntermediate": true
+  },
+  "version": 0.4
+}

ComfyUI-WanBerniniS2V_v2/__init__.py

@@ -0,0 +1,7 @@
+from .model_patch import apply_model_patches
+
+apply_model_patches()
+
+from .nodes import comfy_entrypoint
+
+__all__ = ["comfy_entrypoint"]

ComfyUI-WanBerniniS2V_v2/audio_mask.py

@@ -0,0 +1,68 @@
+import torch
+
+import comfy.utils
+
+WAN_VAE_SCALE = 8
+WAN_PATCH_SPATIAL = 2
+
+
+def _padded_latent_dim(pixels: int) -> int:
+    latent = pixels // WAN_VAE_SCALE
+    return latent + (WAN_PATCH_SPATIAL - latent % WAN_PATCH_SPATIAL) % WAN_PATCH_SPATIAL
+
+
+def token_grid_size(width: int, height: int) -> tuple[int, int]:
+    return _padded_latent_dim(height) // WAN_PATCH_SPATIAL, _padded_latent_dim(width) // WAN_PATCH_SPATIAL
+
+
+def mask_to_token_grid(mask_image: torch.Tensor, width: int, height: int) -> torch.Tensor:
+    token_h, token_w = token_grid_size(width, height)
+    mask = mask_image[0] if mask_image.ndim == 3 else mask_image
+    mask = mask.unsqueeze(0).unsqueeze(0)
+    mask = comfy.utils.common_upscale(mask, width, height, "area", "center")
+    mask = comfy.utils.common_upscale(mask, token_w, token_h, "area", "center")
+    return (mask > 0.5).to(dtype=torch.float32).flatten(2).squeeze(1)
+
+
+def _latent_frame_weight(video_frame: int, start_frame: int, end_frame: int, crossfade_frames: int) -> float:
+    if video_frame < start_frame or video_frame >= end_frame:
+        return 0.0
+    if crossfade_frames <= 0:
+        return 1.0
+    weight = 1.0
+    if video_frame < start_frame + crossfade_frames:
+        weight = min(weight, (video_frame - start_frame + 1) / crossfade_frames)
+    if video_frame >= end_frame - crossfade_frames:
+        weight = min(weight, (end_frame - video_frame) / crossfade_frames)
+    return max(0.0, weight)
+
+
+def build_timeline_audio_inject_mask(
+    width: int,
+    height: int,
+    length: int,
+    segments,
+    crossfade_frames: int = 0,
+    device=None,
+) -> torch.Tensor:
+    latent_t = ((length - 1) // 4) + 1
+    token_h, token_w = token_grid_size(width, height)
+    n_tokens = token_h * token_w
+    mask = torch.zeros(1, latent_t, n_tokens, 1)
+
+    for segment in segments:
+        tokens = mask_to_token_grid(segment["mask_image"], width, height)
+        start_frame = int(segment["start_frame"])
+        end_frame = int(segment["end_frame"])
+        for latent_idx in range(latent_t):
+            vf0 = latent_idx * 4
+            vf1 = vf0 + 4
+            weight = 0.0
+            for video_frame in range(vf0, vf1):
+                weight = max(weight, _latent_frame_weight(video_frame, start_frame, end_frame, crossfade_frames))
+            if weight > 0.0:
+                mask[:, latent_idx, :, 0] = torch.maximum(mask[:, latent_idx, :, 0], tokens * weight)
+
+    if device is not None:
+        mask = mask.to(device)
+    return mask

ComfyUI-WanBerniniS2V_v2/demo_assets/ComfyUI_00207-audio.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a01ab587948bca585beb7422d10cdb7a0cd05dc06740bb97400e8b894e1417ff
+size 1877791

ComfyUI-WanBerniniS2V_v2/demo_assets/I am The Dude Playing The Dude.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1dcc8ec7a4ca3b26196713e65ab214f7a2516778f45d9273829255d098bab2
+size 191604

ComfyUI-WanBerniniS2V_v2/demo_assets/Im the Dude.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4590983cee3e79dda5b59979b84ffa8b811e86c7f19f97483a1fb7ce0c824d97
+size 404056

ComfyUI-WanBerniniS2V_v2/model_patch.py

@@ -0,0 +1,226 @@
+import inspect
+import logging
+
+import torch
+
+import comfy.conds
+import comfy.model_management
+from comfy.ldm.wan.model import AudioInjector_WAN, WanModel_S2V
+from comfy.model_base import WAN22_S2V
+
+
+def _append_context_latents(self, x, kwargs):
+    context_latents = kwargs.get("context_latents", None)
+    if context_latents is None:
+        return x
+    for lat in context_latents:
+        cl = self.patch_embedding(lat.float().to(x.device)).to(x.dtype).flatten(2).transpose(1, 2)
+        x = torch.cat([x, cl], dim=1)
+    return x
+
+
+def _patch_wan_model_s2v_forward():
+    if getattr(WanModel_S2V.forward_orig, "__wan_bernini_s2v_v2_patch__", False):
+        return
+
+    try:
+        source = inspect.getsource(WanModel_S2V.forward_orig)
+    except (OSError, TypeError):
+        source = ""
+    if "context_latents" in source and getattr(WanModel_S2V.forward_orig, "__wan_bernini_s2v_patch__", False):
+        WanModel_S2V.forward_orig.__wan_bernini_s2v_v2_patch__ = True
+        return
+
+    original = WanModel_S2V.forward_orig
+
+    def forward_orig(
+        self,
+        x,
+        t,
+        context,
+        audio_embed=None,
+        reference_latent=None,
+        control_video=None,
+        reference_motion=None,
+        clip_fea=None,
+        freqs=None,
+        transformer_options={},
+        **kwargs,
+    ):
+        if audio_embed is not None:
+            num_embeds = x.shape[-3] * 4
+            audio_emb_global, audio_emb = self.casual_audio_encoder(audio_embed[:, :, :, :num_embeds])
+        else:
+            audio_emb = None
+            audio_emb_global = None
+
+        bs, _, time, height, width = x.shape
+        x = self.patch_embedding(x.float()).to(x.dtype)
+        if control_video is not None:
+            x = x + self.cond_encoder(control_video)
+
+        if t.ndim == 1:
+            t = t.unsqueeze(1).repeat(1, x.shape[2])
+
+        grid_sizes = x.shape[2:]
+        x = x.flatten(2).transpose(1, 2)
+        seq_len = x.size(1)
+
+        cond_mask_weight = comfy.model_management.cast_to(self.trainable_cond_mask.weight, dtype=x.dtype, device=x.device).unsqueeze(1).unsqueeze(1)
+        x = x + cond_mask_weight[0]
+        x = _append_context_latents(self, x, kwargs)
+
+        if reference_latent is not None:
+            ref = self.patch_embedding(reference_latent.float()).to(x.dtype)
+            ref = ref.flatten(2).transpose(1, 2)
+            freqs_ref = self.rope_encode(reference_latent.shape[-3], reference_latent.shape[-2], reference_latent.shape[-1], t_start=max(30, time + 9), device=x.device, dtype=x.dtype)
+            ref = ref + cond_mask_weight[1]
+            x = torch.cat([x, ref], dim=1)
+            freqs = torch.cat([freqs, freqs_ref], dim=1)
+            t = torch.cat([t, torch.zeros((t.shape[0], reference_latent.shape[-3]), device=t.device, dtype=t.dtype)], dim=1)
+
+        if reference_motion is not None:
+            motion_encoded, freqs_motion = self.frame_packer(reference_motion, self)
+            motion_encoded = motion_encoded + cond_mask_weight[2]
+            x = torch.cat([x, motion_encoded], dim=1)
+            freqs = torch.cat([freqs, freqs_motion], dim=1)
+            t = torch.repeat_interleave(t, 2, dim=1)
+            t = torch.cat([t, torch.zeros((t.shape[0], 3), device=t.device, dtype=t.dtype)], dim=1)
+
+        from comfy.ldm.wan.model import sinusoidal_embedding_1d
+
+        e = self.time_embedding(
+            sinusoidal_embedding_1d(self.freq_dim, t.flatten()).to(dtype=x[0].dtype))
+        e = e.reshape(t.shape[0], -1, e.shape[-1])
+        e0 = self.time_projection(e).unflatten(2, (6, self.dim))
+
+        context = self.text_embedding(context)
+
+        patches_replace = transformer_options.get("patches_replace", {})
+        blocks_replace = patches_replace.get("dit", {})
+        transformer_options["total_blocks"] = len(self.blocks)
+        transformer_options["block_type"] = "double"
+        for i, block in enumerate(self.blocks):
+            transformer_options["block_index"] = i
+            if ("double_block", i) in blocks_replace:
+                def block_wrap(args):
+                    out = {}
+                    out["img"] = block(args["img"], context=args["txt"], e=args["vec"], freqs=args["pe"], transformer_options=args["transformer_options"])
+                    return out
+                out = blocks_replace[("double_block", i)]({"img": x, "txt": context, "vec": e0, "pe": freqs, "transformer_options": transformer_options}, {"original_block": block_wrap})
+                x = out["img"]
+            else:
+                x = block(x, e=e0, freqs=freqs, context=context, transformer_options=transformer_options)
+            if audio_emb is not None:
+                inject_scale = kwargs.get("audio_inject_scale", 1.0)
+                if isinstance(inject_scale, torch.Tensor):
+                    inject_scale = inject_scale.reshape(-1)[0].item()
+                x = self.audio_injector(
+                    x, i, audio_emb, audio_emb_global, seq_len,
+                    scale=inject_scale,
+                    token_mask=kwargs.get("audio_inject_mask", None),
+                )
+        x = self.head(x, e)
+        x = self.unpatchify(x, grid_sizes)
+        return x
+
+    forward_orig.__wan_bernini_s2v_v2_patch__ = True
+    forward_orig.__wan_bernini_s2v_patch__ = True
+    forward_orig.__wan_bernini_s2v_original__ = original
+    WanModel_S2V.forward_orig = forward_orig
+
+
+def _patch_audio_injector():
+    if getattr(AudioInjector_WAN.forward, "__wan_bernini_s2v_v2_masked_patch__", False):
+        return
+
+    original_forward = AudioInjector_WAN.forward
+
+    def forward(self, x, block_id, audio_emb, audio_emb_global, seq_len, scale=1.0, token_mask=None):
+        if token_mask is None:
+            return original_forward(self, x, block_id, audio_emb, audio_emb_global, seq_len, scale=scale)
+
+        audio_attn_id = self.injected_block_id.get(block_id, None)
+        if audio_attn_id is None:
+            return x
+
+        from einops import rearrange
+
+        num_frames = audio_emb.shape[1]
+        input_hidden_states = rearrange(x[:, :seq_len], "b (t n) c -> (b t) n c", t=num_frames)
+        if self.enable_adain and self.adain_mode == "attn_norm":
+            audio_emb_global = rearrange(audio_emb_global, "b t n c -> (b t) n c")
+            adain_hidden_states = self.injector_adain_layers[audio_attn_id](input_hidden_states, temb=audio_emb_global[:, 0])
+            attn_hidden_states = adain_hidden_states
+        else:
+            attn_hidden_states = self.injector_pre_norm_feat[audio_attn_id](input_hidden_states)
+
+        if audio_emb.dim() == 3:
+            attn_audio_emb = rearrange(audio_emb, "b t c -> (b t) 1 c", t=num_frames)
+        else:
+            attn_audio_emb = rearrange(audio_emb, "b t n c -> (b t) n c", t=num_frames)
+
+        residual_out = self.injector[audio_attn_id](x=attn_hidden_states, context=attn_audio_emb)
+        residual_out = rearrange(residual_out, "(b t) n c -> b (t n) c", t=num_frames)
+
+        if token_mask.ndim == 4:
+            token_mask = token_mask.flatten(1, 2)
+        if token_mask.shape[1] == residual_out.shape[1]:
+            residual_out = residual_out * token_mask.to(device=residual_out.device, dtype=residual_out.dtype)
+        else:
+            logging.warning(
+                "ComfyUI-WanBerniniS2V_v2: mask length %s does not match token count %s; using global audio injection",
+                token_mask.shape[1],
+                residual_out.shape[1],
+            )
+
+        x[:, :seq_len] = x[:, :seq_len] + residual_out * scale
+        return x
+
+    forward.__wan_bernini_s2v_v2_masked_patch__ = True
+    forward.__wan_bernini_s2v_masked_patch__ = True
+    forward.__wan_bernini_s2v_masked_original__ = original_forward
+    AudioInjector_WAN.forward = forward
+
+
+def _patch_wan22_s2v_extra_conds():
+    if getattr(WAN22_S2V.extra_conds, "__wan_bernini_s2v_v2_masked_patch__", False):
+        return
+
+    original_extra_conds = WAN22_S2V.extra_conds
+
+    def extra_conds(self, **kwargs):
+        out = original_extra_conds(self, **kwargs)
+        audio_inject_mask = kwargs.get("audio_inject_mask", None)
+        if audio_inject_mask is not None:
+            out["audio_inject_mask"] = comfy.conds.CONDRegular(audio_inject_mask)
+        audio_inject_scale = kwargs.get("audio_inject_scale", None)
+        if audio_inject_scale is not None:
+            out["audio_inject_scale"] = comfy.conds.CONDRegular(torch.FloatTensor([audio_inject_scale]))
+        return out
+
+    extra_conds.__wan_bernini_s2v_v2_masked_patch__ = True
+    extra_conds.__wan_bernini_s2v_masked_patch__ = True
+    extra_conds.__wan_bernini_s2v_masked_original__ = original_extra_conds
+    WAN22_S2V.extra_conds = extra_conds
+
+    original_resize = WAN22_S2V.resize_cond_for_context_window
+
+    def resize_cond_for_context_window(self, cond_key, cond_value, window, x_in, device, retain_index_list=[]):
+        if cond_key == "audio_inject_mask":
+            mask = cond_value.cond
+            if mask.ndim == 4 and mask.shape[1] == x_in.shape[2]:
+                return cond_value._copy_with(window.get_tensor(mask, device, dim=1))
+        return original_resize(self, cond_key, cond_value, window, x_in, device, retain_index_list=retain_index_list)
+
+    resize_cond_for_context_window.__wan_bernini_s2v_v2_masked_patch__ = True
+    resize_cond_for_context_window.__wan_bernini_s2v_masked_patch__ = True
+    resize_cond_for_context_window.__wan_bernini_s2v_masked_original__ = original_resize
+    WAN22_S2V.resize_cond_for_context_window = resize_cond_for_context_window
+
+
+def apply_model_patches():
+    _patch_wan_model_s2v_forward()
+    _patch_audio_injector()
+    _patch_wan22_s2v_extra_conds()
+    logging.info("ComfyUI-WanBerniniS2V_v2: applied Bernini S2V model patches")

ComfyUI-WanBerniniS2V_v2/nodes.py

@@ -0,0 +1,153 @@
+import torch
+from typing_extensions import override
+
+import comfy.model_management
+import comfy.utils
+import node_helpers
+from comfy_api.latest import ComfyExtension, io
+
+from .audio_mask import build_timeline_audio_inject_mask
+from .wan_audio import apply_timeline_audio_conditioning, resolve_timeline_segment_ranges
+
+
+def _resize_long_edge(image, max_size, stride=16):
+    h, w = image.shape[1], image.shape[2]
+    scale = min(max_size / max(h, w), 1.0)
+    nh = max(stride, round(h * scale / stride) * stride)
+    nw = max(stride, round(w * scale / stride) * stride)
+    return comfy.utils.common_upscale(image[:, :, :, :3].movedim(-1, 1), nw, nh, "area", "disabled").movedim(1, -1)
+
+
+def _build_context_latents(vae, width, height, length, source_video=None, reference_video=None, reference_images=None, ref_max_size=848):
+    context = []
+    if source_video is not None:
+        vid = comfy.utils.common_upscale(source_video[:length, :, :, :3].movedim(-1, 1), width, height, "area", "center").movedim(1, -1)
+        context.append(vae.encode(vid[:, :, :, :3]))
+
+    if reference_video is not None:
+        ref_vid = _resize_long_edge(reference_video[:length], ref_max_size)
+        context.append(vae.encode(ref_vid[:, :, :, :3]))
+
+    if reference_images:
+        for name in sorted(reference_images):
+            imgs = reference_images[name]
+            if imgs is None:
+                continue
+            for i in range(imgs.shape[0]):
+                img = _resize_long_edge(imgs[i:i + 1], ref_max_size)
+                context.append(vae.encode(img[:, :, :, :3]))
+    return context
+
+
+class BerniniS2VConditioningV2(io.ComfyNode):
+    @classmethod
+    def define_schema(cls):
+        return io.Schema(
+            node_id="BerniniS2VConditioningV2",
+            display_name="Bernini S2V Conditioning v2",
+            category="model/conditioning/bernini",
+            description="Bernini in-context conditioning with masked S2V audio for one or two speakers. Requires a Wan 2.2 S2V grafted Bernini-R model. Paint speaker masks on the output frame; reference_image_0 maps to image0 in prompts.",
+            inputs=[
+                io.Conditioning.Input("positive"),
+                io.Conditioning.Input("negative"),
+                io.Vae.Input("vae"),
+                io.Int.Input("width", default=832, min=16, max=8192, step=16),
+                io.Int.Input("height", default=480, min=16, max=8192, step=16),
+                io.Int.Input("length", default=81, min=1, max=8192, step=4),
+                io.Int.Input("batch_size", default=1, min=1, max=4096),
+                io.AudioEncoderOutput.Input("audio_1"),
+                io.Mask.Input("mask_1", tooltip="White = speaker 1 lip-sync region on the output frame."),
+                io.AudioEncoderOutput.Input("audio_2", optional=True),
+                io.Mask.Input("mask_2", optional=True, tooltip="Required when audio_2 is connected."),
+                io.Int.Input("speaker_2_start_frame", default=-1, min=-1, max=8192, step=1,
+                    tooltip="-1 auto-starts speaker 2 when speaker 1 audio ends."),
+                io.Image.Input("source_video", optional=True),
+                io.Image.Input("reference_video", optional=True),
+                io.Autogrow.Input("reference_images", optional=True,
+                    template=io.Autogrow.TemplatePrefix(
+                        input=io.Image.Input("reference_image"),
+                        prefix="reference_image_", min=0, max=8)),
+                io.Int.Input("ref_max_size", default=848, min=16, max=8192, step=16, optional=True),
+                io.Int.Input("mask_crossfade_frames", default=4, min=0, max=64, step=1,
+                    tooltip="Softens the mask handoff between speakers. 0 = hard cut."),
+                io.Float.Input("audio_inject_scale", default=1.0, min=0.0, max=10.0, step=0.01),
+            ],
+            outputs=[
+                io.Conditioning.Output(display_name="positive"),
+                io.Conditioning.Output(display_name="negative"),
+                io.Latent.Output(display_name="latent"),
+            ],
+        )
+
+    @classmethod
+    def execute(
+        cls,
+        positive,
+        negative,
+        vae,
+        width,
+        height,
+        length,
+        batch_size,
+        audio_1,
+        mask_1,
+        audio_2=None,
+        mask_2=None,
+        speaker_2_start_frame=-1,
+        source_video=None,
+        reference_video=None,
+        reference_images=None,
+        ref_max_size=848,
+        mask_crossfade_frames=4,
+        audio_inject_scale=1.0,
+    ) -> io.NodeOutput:
+        if audio_1 is None:
+            raise ValueError("Bernini S2V Conditioning v2 requires audio_1.")
+        if mask_1 is None:
+            raise ValueError("Bernini S2V Conditioning v2 requires mask_1.")
+        if audio_2 is not None and mask_2 is None:
+            raise ValueError("mask_2 is required when audio_2 is connected.")
+
+        segments = [{"audio_encoder_output": audio_1, "start_frame": 0, "mask_image": mask_1}]
+        if audio_2 is not None:
+            segments.append({
+                "audio_encoder_output": audio_2,
+                "start_frame": speaker_2_start_frame,
+                "mask_image": mask_2,
+            })
+
+        latent = torch.zeros(
+            [batch_size, 16, ((length - 1) // 4) + 1, height // 8, width // 8],
+            device=comfy.model_management.intermediate_device())
+
+        context = _build_context_latents(vae, width, height, length, source_video, reference_video, reference_images, ref_max_size)
+        if context:
+            positive = node_helpers.conditioning_set_values(positive, {"context_latents": context})
+            negative = node_helpers.conditioning_set_values(negative, {"context_latents": context})
+
+        positive, negative = apply_timeline_audio_conditioning(positive, negative, length, segments)
+        resolved_segments = resolve_timeline_segment_ranges(length, segments)
+        cond_values = {
+            "audio_inject_scale": audio_inject_scale,
+            "audio_inject_mask": build_timeline_audio_inject_mask(
+                width, height, length, resolved_segments,
+                crossfade_frames=mask_crossfade_frames,
+                device=comfy.model_management.intermediate_device(),
+            ),
+        }
+        positive = node_helpers.conditioning_set_values(positive, cond_values)
+        negative_values = dict(cond_values)
+        negative_values["audio_inject_mask"] = negative_values["audio_inject_mask"] * 0.0
+        negative = node_helpers.conditioning_set_values(negative, negative_values)
+
+        return io.NodeOutput(positive, negative, {"samples": latent})
+
+
+class WanBerniniS2VV2Extension(ComfyExtension):
+    @override
+    async def get_node_list(self) -> list[type[io.ComfyNode]]:
+        return [BerniniS2VConditioningV2]
+
+
+async def comfy_entrypoint() -> WanBerniniS2VV2Extension:
+    return WanBerniniS2VV2Extension()

ComfyUI-WanBerniniS2V_v2/wan_audio.py

@@ -0,0 +1,89 @@
+import math
+
+import torch
+
+import node_helpers
+from comfy_extras.nodes_wan import get_audio_embed_bucket_fps, linear_interpolation
+
+WAN_AUDIO_INPUT_FPS = 50
+WAN_AUDIO_VIDEO_RATE = 30
+WAN_AUDIO_FPS = 16
+WAN_AUDIO_SAMPLE_RATE = 16000
+
+
+def _audio_feat(audio_encoder_output):
+    feat = torch.cat(audio_encoder_output["encoded_audio_all_layers"])
+    return linear_interpolation(feat, input_fps=WAN_AUDIO_INPUT_FPS, output_fps=WAN_AUDIO_VIDEO_RATE)
+
+
+def audio_encoder_output_video_frames(audio_encoder_output, fps=WAN_AUDIO_FPS):
+    audio_samples = audio_encoder_output.get("audio_samples")
+    if audio_samples is not None:
+        return max(1, int(round(audio_samples / float(WAN_AUDIO_SAMPLE_RATE) * fps)))
+    feat = _audio_feat(audio_encoder_output)
+    return max(1, int(round(feat.shape[1] * fps / WAN_AUDIO_VIDEO_RATE)))
+
+
+def _permute_audio_embed_bucket(audio_embed_bucket):
+    audio_embed_bucket = audio_embed_bucket.unsqueeze(0)
+    if len(audio_embed_bucket.shape) == 3:
+        return audio_embed_bucket.permute(0, 2, 1)
+    return audio_embed_bucket.permute(0, 2, 3, 1)
+
+
+def build_timeline_audio_embed(length, segments):
+    latent_t = ((length - 1) // 4) + 1
+    batch_frames = latent_t * 4
+    total_feat_frames = int(math.ceil(batch_frames * WAN_AUDIO_VIDEO_RATE / WAN_AUDIO_FPS))
+
+    composite = None
+    cursor_auto = 0
+    for segment in segments:
+        feat = _audio_feat(segment["audio_encoder_output"])
+        if composite is None:
+            composite = torch.zeros(
+                feat.shape[0], total_feat_frames, feat.shape[2],
+                dtype=feat.dtype, device=feat.device)
+
+        start_frame = segment.get("start_frame", -1)
+        if start_frame is None or start_frame < 0:
+            start_frame = cursor_auto
+        else:
+            start_frame = int(start_frame)
+
+        start_feat = int(round(start_frame * WAN_AUDIO_VIDEO_RATE / WAN_AUDIO_FPS))
+        copy_len = min(feat.shape[1], total_feat_frames - start_feat)
+        if copy_len > 0 and start_feat < total_feat_frames:
+            composite[:, start_feat:start_feat + copy_len, :] = feat[:, :copy_len, :]
+
+        cursor_auto = start_frame + audio_encoder_output_video_frames(segment["audio_encoder_output"])
+
+    audio_embed_bucket, _ = get_audio_embed_bucket_fps(
+        composite, fps=WAN_AUDIO_FPS, batch_frames=batch_frames, m=0, video_rate=WAN_AUDIO_VIDEO_RATE)
+    audio_embed_bucket = _permute_audio_embed_bucket(audio_embed_bucket)
+    return audio_embed_bucket[:, :, :, :batch_frames]
+
+
+def resolve_timeline_segment_ranges(length, segments):
+    batch_frames = (((length - 1) // 4) + 1) * 4
+    resolved = []
+    cursor_auto = 0
+    for segment in segments:
+        start_frame = segment.get("start_frame", -1)
+        if start_frame is None or start_frame < 0:
+            start_frame = cursor_auto
+        else:
+            start_frame = int(start_frame)
+        end_frame = min(batch_frames, start_frame + audio_encoder_output_video_frames(segment["audio_encoder_output"]))
+        resolved.append({**segment, "start_frame": start_frame, "end_frame": end_frame})
+        cursor_auto = end_frame
+    return resolved
+
+
+def apply_timeline_audio_conditioning(positive, negative, length, segments):
+    audio_embed_bucket = build_timeline_audio_embed(length, segments)
+    if audio_embed_bucket is None or audio_embed_bucket.shape[3] <= 0:
+        return positive, negative
+    positive = node_helpers.conditioning_set_values(positive, {"audio_embed": audio_embed_bucket})
+    negative = node_helpers.conditioning_set_values(negative, {"audio_embed": audio_embed_bucket * 0.0})
+    return positive, negative