search

BadToBest / EchoMimic

Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditioning
https://badtobest.github.io/echomimic.html
Apache License 2.0
2.26k stars 263 forks source link

上传四分钟的音频合出来好像只有前面10s，后面的图片就不会动了 #82

Closed redstoneleo closed 1 month ago

redstoneleo commented 1 month ago

测试平台：https://www.modelscope.cn/studios/BadToBest/BadToBest

O-O1024 commented 1 month ago

可能是 video_length 太小了

JoeFannie commented 1 month ago

调整video_length = fps * time，例如4分钟=240s fps=24，video length调整到240s或者5760帧

redstoneleo commented 1 month ago

这个感觉是可以根据上传的音频时长自动计算的吧，为什么也要用户手动计算后输入呢？