生成结果视频不连续，黑屏

[missu263]

生成结果在12秒时候，就没有画面了。原视频50多秒

使用素材 https://github.com/TMElyralab/MuseTalk/assets/15882147/91062cd4-dc8e-4c11-9357-20d801a93f03

音频：audio_path: "data/audio/sun.wav"

生成结果如下 https://github.com/TMElyralab/MuseTalk/assets/15882147/3b507621-565c-4bde-8711-4e2a9dbfa059

[alexLIUMinhao]

Hi，

Your footage has not been edited well, and it contains many non-human scenes, which may cause the face detection model to fail. Therefore, please prepare a clip that includes frontal human subjects for testing.

[panjie-payne]

Hi，

Your footage has not been edited well, and it contains many non-human scenes, which may cause the face detection model to fail. Therefore, please prepare a clip that includes frontal human subjects for testing.

在有些场景下， 视频中间可能会穿插一些没有人脸的场景， 怎么避免因为出现没有人脸场景导致黑屏的问题呢

[praymich]

Hi， Your footage has not been edited well, and it contains many non-human scenes, which may cause the face detection model to fail. Therefore, please prepare a clip that includes frontal human subjects for testing.

在有些场景下， 视频中间可能会穿插一些没有人脸的场景， 怎么避免因为出现没有人脸场景导致黑屏的问题呢

我也有这个问题，请问你解决了吗

[panjie-payne]

Hi， Your footage has not been edited well, and it contains many non-human scenes, which may cause the face detection model to fail. Therefore, please prepare a clip that includes frontal human subjects for testing.

在有些场景下， 视频中间可能会穿插一些没有人脸的场景， 怎么避免因为出现没有人脸场景导致黑屏的问题呢

我也有这个问题，请问你解决了吗

解决了，自己魔改代码， 查找无效帧，对齐就好了。反正就是如果没有人脸就播放原素材片段就行了