Question about OOD Chinese Dataset

[ndisci]

Hi,

Thanks for this nice work. I have some questions.

1. I have tested your pretrained model on OOD Chinese data you shared with us. I have run gather and preprocess scripts to prepare data then run your predict script . But I couldn't reproduce the results as presented in your paper . I tried it for both device.

2. Where is the ground truth csv file of OOD Chinese data ? You share many files. Preparation and usage of BVCC dataset is so complicated.

3. How about other languages and bonafide wavs ? I test your model on several English datasets and also other language custom data. Results is significantly better than other languages.

Thanks for help .