Jason Cheng
Jason Cheng
> Would it make sense to provide the source package as well on PyPi? Because the following works great on an M1: > > ``` > pip install git+https://github.com/letmaik/pyvirtualcam >...
> Hi, I also get results offset: 4, by running demo script. But I cannot tell what is the unit of the offset. Is it milliseconds? looks like frame index:...
try streaming diffusion.
Windows or linux?
> Hi @XinBow99 may be you can help me, im trying to make inference but im getting some issues in windows may be you know why: the issue is explained...
https://youtu.be/UL6oS8by5VA 1.Voice alignment through algorithms vir virtual device. 2.Currently, frame alignment is necessary to avoid video jitter. I'm trying to fix it. 3.Maybe you can try to combine **forward_system**
看起來跑在2080上約為10fps * 512*512
> 谢谢您，这边解决了，我使用的模型是wav2lip+自己部署的cosyvoice2.0,多个网页请求不同的数字人，大概解决思路原先avartar是一个全局共享变量，现在每个session启动或者切换数字人形象的时候都会独立加载，然后一些推理线程等也是自己加载，ai是这么解决的。 請問現在一次可以推幾路呢