VoiceWukong
收藏资源简介:
VoiceWukong是由华中科技大学创建的一个综合性的深度伪造语音检测基准数据集,旨在评估深度伪造语音检测器的性能。该数据集包含265,200个英语和148,200个中文的深度伪造语音样本,涵盖了19种商业工具和15种开源工具生成的语音。数据集通过六种类型的操作创建了38种数据变体,旨在模拟真实世界中的语音操作。VoiceWukong的应用领域主要集中在深度伪造语音检测,旨在解决当前检测方法在实际应用中的泛化能力不足的问题。
VoiceWukong is a comprehensive deepfake speech detection benchmark dataset created by Huazhong University of Science and Technology, aiming to evaluate the performance of deepfake speech detectors. This dataset contains 265,200 English and 148,200 Chinese deepfake speech samples, covering speech generated by 19 commercial tools and 15 open-source tools. The dataset develops 38 data variants through six types of manipulation operations to simulate real-world speech manipulation scenarios. Its primary application field focuses on deepfake speech detection, and it is designed to address the problem of insufficient generalization capability of current detection methods in practical applications.
VoiceWukong 数据集概述
数据集名称
- VoiceWukong
数据集描述
- 一个综合性的深度伪造语音检测基准。
相关链接
- 排行榜: Leaderboard
- 数据集: Dataset
- 用户研究结果: User Study Results
- 加权模型: Weighted Models
- 原始输出: Original Outputs
- 数据集构建脚本: Dataset Construction Scripts

- 1VoiceWukong: Benchmarking Deepfake Voice Detection华中科技大学 · 2024年



