Utgivere: Norsk Regnesentral
Serie: NR-notat DART/10/22
This is a summary of what is needed to be done to create the models that are used in the SoundVision project. This includes hardware, software, gathering datasets, and workflows for how to create the models. The documentation should make it possible to start from scratch and creating models quickly (although training the models may take some time depending on the hardware).