正在加载视频...
视频加载失败
I've been running some animation tests with Google DeepMind Veo 3's new image2video with voice... If anyone doesn't think that this technology is going to have a profound effect on animation workflows in the coming months they are deluded.... images made with Midjourney and Runway.
11 条评论

@GoogleDeepMind Which video tool will be the first to give us consistently definable sets, characters, mannerisms, props, wardrobes, movable cameras, color grading arcs, voices etc.? I'd love for my tools' dashboard to basically be my current visual worldbuilding document. A dollhouse!

@GoogleDeepMind That's certainly what a lot of research is going into at present.

@GoogleDeepMind As soon as you can save the voice selection it gives as a reference to reuse its all gravy 🙌 Would voice change with elevenlabs keep the pacing to put over your vids?

@GoogleDeepMind Tried the elevenlabs voice changer trick and it kills a lot of performance in the voice.

@GoogleDeepMind Yeah the convenience of doing animation and lip sync at the same time is sooooo nice! Anything without seems archaic now. I hope more platforms figure this out, seems like Runway may be attempting to with a cryptic post from Cristobal

@GoogleDeepMind I think they are cooking a few things at Runway....

@GoogleDeepMind Runway Gen 4 image référence+ veo 3 🔥🔥🔥🤯

@GoogleDeepMind Absolutely agree—this is a game-changer. The speed and quality Veo 3 delivers will completely reshape animation pipelines. We’re witnessing the future of storytelling unfold in real time.

@GoogleDeepMind 💯

@GoogleDeepMind Legit 👌

@GoogleDeepMind Cheers Boss.
