x-lance / anitalker Goto Github PK

Training and Inference Hardware.
Specifically, the GPU with 8G
VRAM can generate up to 3 minutes of video in one inference.

Hi there,

First of all, splendid work! I really appreciate the generalizability of this model and am looking forward to the code release.

I have a question regarding the paper's mention that a single 8GB GPU can produce a 3-minute video. Is it possible to make this process real-time or to stream the generated video so it appears as a real-time conversation?

Which kind of GPU should I use? 4090 or just mac M1?

I didn't find the code folder in README.md?

You mention a Model Zoo checkpoint required to make this work, but there are no links to any checkpoints nor any specification of which checkpoint must be downloaded.

You mention a Model Zoo checkpoint required to make this work, but there are no links to any checkpoints nor any specification of which checkpoint must be downloaded.

你好，非常感谢您的工作，想问一下代码什么时候可以发布

Will you release the code for loss function L_{MI} and L_{ML}, which are the core of this paper, please?