DCASE2020_Task2_Solution-Anomaly_detection-
DCASE2020_Task2_Solution-Anomaly_detection- copied to clipboard
shinmura0
how to get the values of max_,min_,mean and sigma of the dataset?
@shinmura0 Hi. how to get the values of max_,min_,mean and sigma of the dataset? max_ = np.max(x_train)=0.18673706 min_ = np.min(x_train)=-0.18011475 But they are different from yours(max_ = 15.6,min_ = -74.7) Also how to get mean and sigma ? Thanks!
This code contains normalization. But this normalization don't contains single machine(or id). This contains other machine(18 id).
@shinmura0 Hi. how to get the values of max_,min_,mean and sigma of the dataset? max_ = np.max(x_train)=0.18673706 min_ = np.min(x_train)=-0.18011475 But they are different from yours(max_ = 15.6,min_ = -74.7) Also how to get mean and sigma ? Thanks!
Hello, I would like to ask asvspoof 2019 PA against recording attack, I repeat the result eval erer:2.0 ( seresnet34 model). I used librosa tool to extract features.but the original paper is based on Kaldi. There are some problems with the evaluation indicators eer. If there is a run through code is better, see which side of me is different from you
@wangtao2668129173
asvspoof 2019 PA? Kaldi? What are you talking about? A competition about the anti replay, I learned that it reproduced an article about senet34 （ASSERT: Anti-Spoofing with Squeeze-Excitation and Residual neTworks）used to predict whether it was a real voice or a repaly