some instances of the input parameters.

[zhangshengnan96]

Sorry to disturb you, I am very new of ML, and I try to run the demo at first, but I don't know the input parameters except the modle path, should be like. So could just give some instances of these input parameters?

[sky1456723]

Sorry for my late reply.
Actually, in MBNet's paper, they do not use bias net during testing as the valid function in test.py. But validation may require it, so I do not remove it directly.
For a quick demo, you can load the model under 'pre_trained/', set step=0, and change the "valid" parameter in line 72 and line 75 of test.py to False. Then it will generate an id table automatically. Otherwise, you can also run train.py, it will generate an id table, too.
You should also download the VCC2018 dataset and specify the data_path parameter of test.py. For example, I can download it and put all the .wav files under data/.