xxuan-speech/Deepfake-Eval-2024-Protocals

Deepfake-Eval-2024是一个用于语音深度伪造检测的数据集，包含训练集、验证集和测试集，具体数据量如下：训练集共28620条数据，其中真实语音18582条，伪造语音10038条；验证集共3179条数据，其中真实语音2059条，伪造语音1120条；测试集共21347条数据，其中真实语音13325条，伪造语音8022条。Deepfake-Eval-2024 is a dataset for speech deepfake detection, containing training, validation, and test sets. The specific data volume is as follows: the training set has a total of 28,620 pieces of data, including 18,582 real speeches and 10,038 fake speeches; the validation set has a total of 3,179 pieces of data, including 2,059 real speeches and 1,120 fake speeches; the test set has a total of 21,347 pieces of data, including 13,325 real speeches and 8,022 fake speeches.