sky1456723 / pytorch-mbnet Goto Github PK

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?

Please provide a demo code of how to create judge ids if anyone wants to retrain the model on their own dataset.

Hi, thanks for the implementation!
I doubt if it is reasonable to use ReLU in both mean_net_dnn and bias_net_dnn. It might be reasonable for mean_net_dnn because the MOS value range is {1,...,5}, but I think we should allow negative bias values. Maybe tanh would be a better choice?

Is it okay to provide the csv file of VCC 2016 dataset for calculating the VCC 2016 system-level LCC and SRCC and MSE?

Hi,

In the paper MBNET: MOS PREDICTION FOR SYNTHESIZED SPEECH WITH MEAN-BIAS NETWORK, the author said the frame-level loss borrowed form MOSNet was used. But I can't find it in your code. I reproduced the frame-level loss and found that the effect is not good. So I want to ask if you have used this loss and how effective it is.

Liu