zh-CN/tutorials/video/wan/wan2-2-s2v #409
Replies: 9 comments 11 replies
-
| 
         wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         大佬求助,如果我用一个3s的音频文件怎么控制视频生成也是3s,可以直接控制length吗  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         音频编码器,有没有中文版本的 wav2vec2_large_chinese_fp16.safetensors 的模型可以下载?要safetensors格式哦  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         下载demo的workflow提示node type:866cdb67-cb97-4a54-8738-e9ece4f5f25f不存在呢  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         最新版官方工作流中WanSoundImageToVideo有bug。输出的cfg为非浮点数,输出的length为非整数。 
 
  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         这个只能生成16帧每秒的帧率么,如果我想生成25帧每秒的,我需要改什么  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         你好,Tesla V100 16G显卡支持fp8吗?  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         你好,希望生成28秒视频,在Template基础上又加了三个“Video S2V Extend subgraph nodes”, batch size设为6,但还是只能生成14秒视频。请问是少了什么步骤吗?谢谢  | 
  
Beta Was this translation helpful? Give feedback.
-
| 
         请问一下,在我一共增加30个“Video S2V Extend”扩展模块,扩展模块采样计算完毕后,VEA解码时,出现一下错误提示:“VAE decoding. Allocation on device This error means you ran out of memory on your GPU. TIPS: If the workflow worked before you might have accidentally set the batch_size to a large number. show report. help fix this” 我的理解是VEA解码时候,显存大小不足导致溢出错误。请问除了减少扩展模块数量这个办法外,还有其他办法解决这个错误吗。(因为我希望能一次性生成足够长的视频来匹配音频时长,我用的显卡是2080TI 11G版本)  | 
  
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
zh-CN/tutorials/video/wan/wan2-2-s2v
这是一个基于 ComfyUI 的 Wan2.2-S2V 音频驱动视频生成原生工作流示例。
https://docs.comfy.org/zh-CN/tutorials/video/wan/wan2-2-s2v
Beta Was this translation helpful? Give feedback.
All reactions