- Execute scripts/setup-project.sh. This will populate the files from the AWS original repo and the ml-source-fsd50k model files from the models directory.
- Open the STM32CubeIDE. When prompted for the workspace path navigate to the stm32 directory. Note that the workspace has to be located there, or otherwise dependencies will not work correctly. The stm32 directory name indirection is preserved from the original project in order to make easy transition to the AWS AI and build framework.
- Select File -> Open Projects From File System from the menu. Navigate to the stm32 directory in this repo and click Open.
- Uncheck all folders that appear in the Folders list and leave the stm32/Projects/b_u585i_iot02s_ntz directory checked.
- Click Finish.