BiGy2018

Results 5 comments of BiGy2018

I write a blog for real-time Wav2Lip implementation: https://medium.com/@bigy2020real/a-holiday-experiment-developing-a-real-time-digital-human-interface-for-llms-ff2e7f3ebc8a

For SVHN->MNIST

> nerfreal.py最后几行注释打开试试: delay = 0.04 - (time.time() - t) #40ms if delay > 0: time.sleep(delay) 感谢,这样正常了。但如果想让fps=50需要怎么做呢?我改了几个地方,但似乎inference卡住了: ![image](https://github.com/lipku/metahuman-stream/assets/27851064/5685c409-a5d3-4e52-8e13-49549a2ef7ce) ![image](https://github.com/lipku/metahuman-stream/assets/27851064/c2e1051c-714b-4dd6-9ac0-e39f677a2d02) ![image](https://github.com/lipku/metahuman-stream/assets/27851064/73a8adeb-1b46-4c49-901e-a850591fd9f1)

> > 这样正常了。但如果想让fps=50需要怎么做呢?我改了几个地方,但似乎inference卡住了: > > 您好,请问您这个说话的界面是从哪出来的,我运行了docker, app.py也能正常运行,但是vlc拉流拉不到,没显示界面,提取感谢指导! rtmpserver运行了么。你需要把app.py的流推到rtmpserver,也就是另一个docker container。注意保证rtmpserver和app.py运行的docker container在同一个网关下。