re-use-demo
An unofficial demo that runs a speech enhancement model in the browser with WebGPU.
What does it do?
This demo runs NVIDIA NemotronLabs' RE-USE speech enhancement model locally in the browser. The model aims to reduce noise and echo mixed with speech and sound problems caused by poor microphones. The provided page description does not explain the specific input method or how to save output files.
Key features
- Run the model locally in the browser using WebGPU
- Improve speech affected by noise, echo, or poor microphone sound