I’ve been building SignBridge, a prototype that takes a word and generates a video of someone signing it in ASL.
See the SignBridge demo: https://whysignbridge.framer.website I’ve spent years learning ASL and participating in the Deaf community. That led me to explore whether generative video could make individual signs easier to access.
What works so far: The prototype can generate videos for individual words. I’ve experimented with prompts that name the ASL sign directly and with more detailed instructions for handshape and movement.
What I’m stuck on: A video can look convincing while getting the handshape, position, or motion wrong. Results also vary between generations. Sentences are a much harder problem, so I’m focusing on single words for now.
If you’ve worked on pose estimation, animation, video generation, or sign language technology, I’d love your perspective: how would you make the hand movements more consistent and verifiable? I’m especially interested in approaches that can be checked with Deaf signers rather than relying on a video looking plausible.