top of page

Experimenting with runwayML Gen3 video-to-video



I can't stop playing with Runway Gen3 video-to-video tool. After my last post, I figured I'd share some more thoughts and another clip.


The 10-second video limit is pretty brutal but there is an option to extend it to 40 seconds after the original clip is made. The tech is moving so fast that this probably won't be an issue for much longer.


Runway's existing prompts are pretty useful, but it's hard to get specific and create completely different prompt approaches and still keep sign language crystal clear. Again, this is likely a short-term issue that will be fixed fairly quickly in the grand scheme of things. 


GoSign.AI is still waiting on the API for Gen3 video-to-video, but we're already diving into Runway's API in general to find ways to bring this tech over to our mobile app. 


Sometimes the GoSign.AI team just wants to add features to our app that aren't just about improving accessibility, but about empowering our users to have fun as well. We're all about leveraging tools like Runway for that purpose.


This way, we treat GoSign.AI mobile app as a testbed for these technologies without having to rely on strict standards for meeting communication needs or educational standards. Since it's just for fun, there's less of a barrier to entry, and we get to test new technologies at a faster pace.



If you haven't given Runway a try yet, check it out: https://runwayml.com/




bottom of page