Download SDK v0.5 (Beta) Documentation GitHub Brand Guidelines
You may download the SDK from the above link, and consult our documentation and visit our GitHub page for API headers and samples. For support please email us at: broadcastsdk-support@nvidia.com
Key Features
- Noise Removal
- Supports wideband (16 kHz) and full band (48 kHz) speech
- 18 audio profiles supported
- AI Accelerated using Tensor Cores
Requirements
Operating System |
Windows 10 64-bit |
Dependencies |
NVIDIA GeForce RTX 2060, Quadro RTX 3000, TITAN RTX or higher
NVIDIA Display Driver 456.38 or more recent
|
Development Environment |
Microsoft Visual Studio 2015 (MSVC14.0) or later
CMake 3.9 or above |
Release Notes
v0.5 (Beta):
- Tripled the number of noise removal profiles based on the feedback from the community beta (codenamed RTX Voice).
- Better noise removal and improved quality for retained speech (improved from RTX Voice beta)
- Support for wideband and fullband speech.
- Support for GeForce RTX 30 and RTX 20 Series, TITAN RTX, and Quadro RTX GPUs
- Easy-to-use C-API