search

KylinYee / R2-Talker-code

R2-Talker: Realistic Real-Time Talking Head Synthesis with Hash Grid Landmarks Encoding and Progressive Multilayer Conditioning
MIT License
71 stars 4 forks source link

Landmark Generator #6

Open KhalilWong opened 5 months ago

KhalilWong commented 5 months ago

If I need to use a custom audio wav file to synthesize a video, is it only possible with "landmark generator"? Also, when is the "landmark generator" expected to be released? Thanks!

KylinYee commented 5 months ago

Yes, you must use the landmark generator. We plan to release the landmark generator before March. If necessary, you can temporarily use the landmark generator in the geneface .

KhalilWong commented 5 months ago

Yes, you must use the landmark generator. We plan to release the landmark generator before March. If necessary, you can temporarily use the landmark generator in the geneface .

How can i use the landmark generator in the geneface with audio rather than video as its input?

xiaoxiongzhg commented 4 months ago

Yes, you must use the landmark generator. We plan to release the landmark generator before March. If necessary, you can temporarily use the landmark generator in the geneface .

Hello author, I know it’s a bit offensive to ask this, when will the code be released?

kike-0304 commented 3 months ago

是的，您必须使用地标生成器。我们计划在三月份之前发布地标生成器。如果需要，您可以暂时使用geneface中的地标生成器。

我如何在geneface中使用音频而不是视频作为输入的地标生成器？

请问您解决这个问题了吗，音频怎么用geneface生成landmark？

KhalilWong commented 3 months ago

还没有，再等等看吧