Download
Implementation of NWT, audio-to-video generation, in Pytorch. The paper proposes a new discrete latent representation named Memcodes, which can be succinctly described as a type of multi-head hard-attention to learned memory (codebook) key/values. They claim the need for less codes and smaller codebook dimensions in order to achieve better reconstructions.
Features
- Implementation of NWT
- Audio-to-video generation
- For Pytorch
- Multi-head hard-attention to learned memory (codebook) key / values
- Smaller codebook dimension
- Achieve better reconstructions
Project Samples
License
MIT LicenseFollow NWT - Pytorch (wip)
Other Useful Business Software
Auth0 for AI Agents now in GA
Connect your AI agents to apps and data more securely, give users control over the actions AI agents can perform and the data they can access, and enable human confirmation for critical agent actions.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of NWT - Pytorch (wip)!
