Dataset_ASVspoof2021_eval 返回的为什么是x_inp, key 而不是返回x_inp, label[key]

[Liuyang0001]

class Dataset_ASVspoof2021_eval(Dataset):
    def __init__(self, list_IDs, base_dir):
        """
            self.list_IDs	: list of strings (each string: utt key),
        """

        self.list_IDs = list_IDs
        self.base_dir = base_dir

    def __len__(self):
        return len(self.list_IDs)

    def __getitem__(self, index):
        self.cut = 64600  # take ~4 sec audio (64600 samples)
        key = self.list_IDs[index]
        X, fs = librosa.load(self.base_dir + 'flac/' + key + '.flac', sr=16000)
        X_pad = pad(X, self.cut)
        x_inp = Tensor(X_pad)
        return x_inp, key

[TonyWangX]

返回的为什么是x_inp, key 而不是返回x_inp, label[key]

Well, please write the question in English since not all the contributors of this repo reads Chinese : )

The question is why getitem returns x_inp, key, rather than x_inp, label[key]

I will send the question to the author of this code.

[TakHemlata]

Hello,

In the given code getitem returns x_inp, key where key is utterance ID "LA_E_000001". You can't return label[key] which is 0 and 1 because in the challenge ground truths (labels) are not provided by Organisers.

Now the labels for 2021 challenge are publicly available so you can return label[key].

Thanks

[shizhuoyang]

Hello,

In the given code getitem returns x_inp, key where key is utterance ID "LA_E_000001". You can't return label[key] which is 0 and 1 because in the challenge ground truths (labels) are not provided by Organisers.

Now the labels for 2021 challenge are publicly available so you can return label[key].

Thanks

Excuse me, you said the labels for 2021 challenge are publicly available, I wonder where it is, could you please give me a link? I download the 2021 dataset from zenodo, which only includes audio.

[TakHemlata]

Hello,

you can download it from the ASVspoof website (https://www.asvspoof.org/)

1st December:
Keys and metadata are now available: LA, PA and DF

Thanks