工作流 /
视频 /
同步3:处理图像

同步3:处理图像
Sync 3: Talking Image
从单张面部照片和一段音频CLIP生成一个说话头像视频,嘴唇和下半部分面部动作与语音精确同步。此过程从一张图像和一个音频输入输出一个视频,保持手部、产品和背景完全不变。适用于产品演示、电商介绍和营销视频,其中静态的产品持有照片与配音配对。
视频
5 个节点 2 个输入 12 次使用 ComfyUI ≥ 0.28.0 ComfyUI 4 KB
节点构成
LoadAudio
× 1
RecordAudio
× 1
SaveVideo
× 1
SyncTalkingImageNode
× 1
LoadImage
× 1
官方标注的模型
Sync 3
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "20a94987-05bd-45a1-9b53-70c45b95a30b",
3 "revision": 0,
4 "last_node_id": 15,
5 "last_link_id": 10,
6 "nodes": [
7 {
8 "id": 11,
9 "type": "LoadAudio",
10 "pos": [
11 -825.2593029405318,
12 639.7313815012487
13 ],
14 "size": [
15 290,
16 160
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "AUDIO",
25 "type": "AUDIO",
26 "links": [
27 8
28 ]
29 }
30 ],
31 "properties": {
32 "cnr_id": "comfy-core",
33 "ver": "0.27.0",
34 "Node name for S&R": "LoadAudio"
35 },
36 "widgets_values": [
37 "gel_cream_ad_voice.mp3",
38 null,
39 null
40 ]
41 },
42 {
43 "id": 12,
44 "type": "RecordAudio",
45 "pos": [
46 -821.2525958894892,
47 872.7020359474577
48 ],
49 "size": [
50 277.5,
51 136.78125
52 ],
53 "flags": [],
54 "order": 1,
55 "mode": 4,
56 "inputs": [],
57 "outputs": [
58 {
59 "name": "AUDIO",
60 "type": "AUDIO",
61 "links": null
62 }
63 ],
64 "properties": {
65 "cnr_id": "comfy-core",
66 "ver": "0.27.0",
67 "Node name for S&R": "RecordAudio"
68 },
69 "widgets_values": [
70 "",
71 ""
72 ]
73 },
74 {
75 "id": 13,
76 "type": "SaveVideo",
77 "pos": [
78 -73.49985253114892,
79 232.59387976105648
80 ],
81 "size": [
82 350,
83 628.65625
84 ],
85 "flags": [],
86 "order": 4,
87 "mode": 0,
88 "inputs": [
89 {
90 "name": "video",
91 "type": "VIDEO",
92 "link": 9
93 }
94 ],
95 "outputs": [
96 {
97 "name": "video",
98 "type": "VIDEO",
99 "links": null
100 }
101 ],
102 "properties": {
103 "cnr_id": "comfy-core",
104 "ver": "0.27.0",
105 "Node name for S&R": "SaveVideo"
106 },
107 "widgets_values": [
108 "video/lip_sync_3.0",
109 "auto",
110 "auto"
111 ]
112 },
113 {
114 "id": 14,
115 "type": "SyncTalkingImageNode",
116 "pos": [
117 -494.8631888682577,
118 233.84789095907297
119 ],
120 "size": [
121 395.296875,
122 328
123 ],
124 "flags": [],
125 "order": 3,
126 "mode": 0,
127 "showAdvanced": false,
128 "inputs": [
129 {
130 "name": "image",
131 "type": "IMAGE",
132 "link": 10
133 },
134 {
135 "name": "audio",
136 "type": "AUDIO",
137 "link": 8
138 }
139 ],
140 "outputs": [
141 {
142 "name": "VIDEO",
143 "type": "VIDEO",
144 "links": [
145 9
146 ]
147 }
148 ],
149 "properties": {
150 "cnr_id": "comfy-core",
151 "ver": "0.28.0",
152 "Node name for S&R": "SyncTalkingImageNode"
153 },
154 "widgets_values": [
155 "Close-up beauty ad. She holds the Berry Gel Cream tube near her cheek, looks at camera, and speaks softly with subtle lip movement. Keep mouth opening small and natural, no exaggerated wide open mouth. Gentle smile, calm head motion, about 5–6 seconds.",
156 1168713874,
157 "randomize",
158 "sync-3",
159 "default",
160 0,
161 0,
162 true
163 ],
164 "color": "#432",
165 "bgcolor": "#653"
166 },
167 {
168 "id": 15,
169 "type": "LoadImage",
170 "pos": [
171 -825.5582673745101,
172 233.03775713554364
173 ],
174 "size": [
175 282.796875,
176 340
177 ],
178 "flags": [],
179 "order": 2,
180 "mode": 0,
181 "inputs": [],
182 "outputs": [
183 {
184 "name": "IMAGE",
185 "type": "IMAGE",
186 "links": [
187 10
188 ]
189 },
190 {
191 "name": "MASK",
192 "type": "MASK",
193 "links": null
194 }
195 ],
196 "properties": {
197 "cnr_id": "comfy-core",
198 "ver": "0.28.0",
199 "Node name for S&R": "LoadImage"
200 },
201 "widgets_values": [
202 "girl_holding_gel_cream.png",
203 "image"
204 ]
205 }
206 ],
207 "links": [
208 [
209 8,
210 11,
211 0,
212 14,
213 1,
214 "AUDIO"
215 ],
216 [
217 9,
218 14,
219 0,
220 13,
221 0,
222 "VIDEO"
223 ],
224 [
225 10,
226 15,
227 0,
228 14,
229 0,
230 "IMAGE"
231 ]
232 ],
233 "groups": [],
234 "config": [],
235 "extra": {
236 "ds": {
237 "scale": 0.8622868016514477,
238 "offset": [
239 1003.0921295331692,
240 52.46606848557645
241 ]
242 },
243 "frontendVersion": "1.45.21"
244 },
245 "version": 0.4
246}