官方服务：
资源简介：
Deepfake Audio Detection Dataset
应用场景：
创建时间：
2026-01-22
相关数据集
STOPA
STOPA数据集是一个公开的语音深度伪造源追踪数据库，包含699,000个来自13个攻击系统的语音样本。该数据集用于训练和评估源追踪模型，旨在解决语音深度伪造的源追踪问题。数据集由13种攻击系统的语音样本组成，包括8种声学模型和6种声码器模型。每个语音样本都被标记了攻击ID、声学模型ID和声码器模型ID，这使得多级源追踪成为可能。
arXiv2025-09-29 更新410
A Relative Phase based Audio Integrity Protection Method-Model and Strategy
Some available data
DataCite Commons2020-08-29 更新50
DeepFense/SpeechFake
--- license: apache-2.0 --- # SpeechFake Dataset Please use the download scripts from https://github.com/XIAOYixuan/AUDDT/tree/yixuan-dev to download and process the dataset. ``` chmod +x download/
Hugging Face2025-12-27 更新50
UCI Capstone: All Real and Generated Audios in MP3 Format
Contains ~22k real audio from the FMA medium dataset and 18k generated audio from the FakeMusicCaps dataset. All audios are preprocessed in .mp3 format. Sampling rate is 16k Hz and bit rate is 215 kb
Zenodo2025-03-05 更新60
UCI Capstone: All Real and Generated Audios in MP3 Format
Contains ~22k real audio from the FMA medium dataset and 18k generated audio from the FakeMusicCaps dataset. All audios are preprocessed in .mp3 format. Sampling rate is 16k Hz and bit rate is 215 kb/
NIAID Data Ecosystem50
