Attacker Attribution of Audio Deepfakes

计算机科学 误传 领域(数学) 归属 领域(数学分析) 计算机安全 人工智能 块(置换群论) 机器学习 几何学 心理学 数学 社会心理学 数学分析 纯数学
作者
Nicolas M. Müller,Franziska Dieckmann,J. L. Williams
出处
期刊:Cornell University - arXiv
标识
DOI:10.48550/arxiv.2203.15563
摘要

Deepfakes are synthetically generated media often devised with malicious intent. They have become increasingly more convincing with large training datasets advanced neural networks. These fakes are readily being misused for slander, misinformation and fraud. For this reason, intensive research for developing countermeasures is also expanding. However, recent work is almost exclusively limited to deepfake detection - predicting if audio is real or fake. This is despite the fact that attribution (who created which fake?) is an essential building block of a larger defense strategy, as practiced in the field of cybersecurity for a long time. This paper considers the problem of deepfake attacker attribution in the domain of audio. We present several methods for creating attacker signatures using low-level acoustic descriptors and machine learning embeddings. We show that speech signal features are inadequate for characterizing attacker signatures. However, we also demonstrate that embeddings from a recurrent neural network can successfully characterize attacks from both known and unknown attackers. Our attack signature embeddings result in distinct clusters, both for seen and unseen audio deepfakes. We show that these embeddings can be used in downstream-tasks to high-effect, scoring 97.10% accuracy in attacker-id classification.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
京京发布了新的文献求助10
2秒前
bkagyin应助wxy采纳,获得10
2秒前
3秒前
qin完成签到,获得积分10
3秒前
3秒前
SCINEXUS应助霸气小懒虫采纳,获得10
3秒前
5秒前
5秒前
6秒前
倪小呆发布了新的文献求助10
6秒前
852应助hhy采纳,获得10
6秒前
大个应助LY采纳,获得10
6秒前
顾矜应助lh采纳,获得10
8秒前
舒心的亦瑶完成签到,获得积分10
8秒前
柳絮发布了新的文献求助10
8秒前
隔壁小曾发布了新的文献求助10
8秒前
彭于晏应助AIIYL采纳,获得10
9秒前
勤劳菠萝发布了新的文献求助10
10秒前
10秒前
11秒前
12秒前
suqinqin发布了新的文献求助10
12秒前
13秒前
14秒前
陈杰完成签到,获得积分10
15秒前
poiuy发布了新的文献求助10
15秒前
彦君发布了新的文献求助10
17秒前
poorzz发布了新的文献求助20
17秒前
18秒前
qq完成签到,获得积分10
18秒前
没头脑完成签到,获得积分10
19秒前
含蓄惊蛰完成签到,获得积分10
19秒前
LY发布了新的文献求助10
20秒前
22秒前
gjww应助朝阳采纳,获得10
23秒前
Hus11221完成签到,获得积分10
24秒前
Daddybo完成签到,获得积分10
24秒前
Cker发布了新的文献求助20
25秒前
25秒前
高分求助中
The three stars each : the Astrolabes and related texts 1070
Manual of Clinical Microbiology, 4 Volume Set (ASM Books) 13th Edition 1000
Hieronymi Mercurialis Foroliviensis De arte gymnastica libri sex: In quibus exercitationum omnium vetustarum genera, loca, modi, facultates, & ... exercitationes pertinet diligenter explicatur Hardcover – 26 August 2016 900
Sport in der Antike 800
De arte gymnastica. The art of gymnastics 600
Sport in der Antike Hardcover – March 1, 2015 500
Boris Pesce - Gli impiegati della Fiat dal 1955 al 1999 un percorso nella memoria 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 有机化学 工程类 生物化学 纳米技术 物理 内科学 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 电极 光电子学 量子力学
热门帖子
关注 科研通微信公众号,转发送积分 2403390
求助须知:如何正确求助?哪些是违规求助? 2102336
关于积分的说明 5304757
捐赠科研通 1829944
什么是DOI,文献DOI怎么找? 911923
版权声明 560458
科研通“疑难数据库(出版商)”最低求助积分说明 487581