Generalization of Audio Deepfake Detection

Chen, Tianxiang; Kumar, Avrosh; Nagarsheth, Parav; Sivaraman, Ganesh; Khoury, Elie

doi:10.21437/odyssey.2020-19

Cited by 124 publications

(71 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A summary of existing approaches can be found in the recent ASVspoof challenges [3,4]. While some of the methods from this field, such as those in [5,6,7], may be applicable to the problem at hand, they typically address a different class of voice modification such as voice conversion or voice synthesis aiming at triggering a false positive voice verification. A key difference with these works, and with other ASV spoof approaches, is the common assumption that speech samples, or features of speech samples, from the original speaker are available; which is not the case for our problem.…”

Section: Relation To Prior Workmentioning

confidence: 99%

On the Detection of Pitch-Shifted Voice: Machines and Human Listeners

Looney¹,

Gaubitch²

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We present a performance comparison between human listeners and a simple algorithm for the task of speech anomaly detection. The algorithm utilises an intentionally small set of features derived from the source-filter model, with the aim of validating that key components of source-filter theory characterise how humans perceive anomalies. We furthermore recognise that humans are adept at detecting anomalies without prior exposure to a given anomaly class.To that end, we also consider the algorithm performance when operating via the principle of unsupervised learning where a null model is derived from normal speech recordings. We evaluate both the algorithm and human listeners for pitch-shift detection where the pitch of a speech sample is intentionally modified using software, a phenomenon of relevance to the fields of fraud detection and forensics.Our results show that humans can only detect pitch-shift reliably at more extreme levels, and that the performance of the algorithm matches closely with that of humans.

show abstract

Section: Relation To Prior Workmentioning

confidence: 99%

On the Detection of Pitch-Shifted Voice: Machines and Human Listeners

Looney¹,

Gaubitch²

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Led by the ASVspoof initiative and corresponding challenge series, countermeasure (CM) systems have hence been developed in order to help detect and deflect spoofing attacks [8][9][10]. In the case of a logical access, telephony scenario involving only TTS and VC attacks, the best performing spoofing CM systems can deliver EERs of less than 2% [11][12][13][14][15][16][17][18][19][20][21].…”

Section: Introductionmentioning

confidence: 99%

SASV Challenge 2022: A Spoofing Aware Speaker Verification Challenge Evaluation Plan

Jung¹,

Tak²,

Shim³

et al. 2022

Preprint

View full text Add to dashboard Cite

“…Hence, tackling these attacks through means of efficient features and machine learning algorithms are a desideratum. The studies in anti-spoofing or countermeasures have increased tremendously with increasing attacks on main-frame systems such as phone-banking theft, unauthentic access to workplaces or even smart phone devices where speech is used as the identity [3], [4]. So, as authentication is no more limited to finger prints and retina scans, the speech based spoofing attacks are growing and catching attention of many researchers for developing robust spoofing detection schemes.…”

Section: Introductionmentioning

confidence: 99%

A Unique Glottal Flow Parameters based Features for Anti-spoofing Countermeasures in Automatic Speaker Verification

Chadha¹,

Abdullah²,

Angeline³

2021

IJACSA

View full text Add to dashboard Cite

The domain of Automatic Speaker Verification (ASV) is blooming with growing developments in feature engineering and artificial intelligence. Inspite of this, the system is liable to spoofing attacks in the form of synthetic or replayed speech. The difficulty in detecting synthetic speech is due to recent advancements in the Voice conversion and Text-to-speech systems which produce natural, indistinguishable speech. To prevent such attacks, there is a need to develop robust spoof detection systems. In order to achieve this goal, we are proposing estimation of Glottal Flow Parameters (GFP) from speech of genuine speech and synthetic spoof samples. The GFP are further parameterized using time, frequency and Liljencrants-Fant (LF) models. Along with GFP features, the Linear Prediction Cepstrum Co-efficient (LFCC) and statistical parameters are computed. The GFP features are investigated to prove their usefulness in detecting spoofed and genuine speech. The ASV spoof 2019 corpus is used to test the framework and evaluated against the baseline models. The proposed spoof detection framework produces an Equal Error Rate (EER) of 2.39% and tandem Detection Cost Function (t-DCF) of 0.0562 which is found to be better than the state-of-the art technique.

show abstract

Generalization of Audio Deepfake Detection

Cited by 124 publications

References 7 publications

On the Detection of Pitch-Shifted Voice: Machines and Human Listeners

On the Detection of Pitch-Shifted Voice: Machines and Human Listeners

SASV Challenge 2022: A Spoofing Aware Speaker Verification Challenge Evaluation Plan

A Unique Glottal Flow Parameters based Features for Anti-spoofing Countermeasures in Automatic Speaker Verification

Contact Info

Product

Resources

About