For slightly faster inference (Neural speech recognition and processing), NVIDIA GPU owners can utilize the CUDA toolkit.
DRE automatically downloads and installs the required runtime library for using CUDA. Should this not work as intended, here’s how to install CUDA yourself:
If you have an NVIDIA GPU, you may download and install CUDA 13.0.2, to decrease the time from speech to command a bit:

During installation, choosing Custom (Advanced) installation and selecting only the CUDA -> Runtime components should be sufficient.
Note! DRE uses Vulkan runtime by default which supports both NVIDIA and AMD GPUs, meaning you do not have to switch to CUDA.