2017
DOI: 10.1007/978-3-319-66429-3_16
|View full text |Cite
|
Sign up to set email alerts
|

Audio-Replay Attack Detection Countermeasures

Abstract: This paper presents the Speech Technology Center (STC) replay attack detection systems proposed for Automatic Speaker Verification Spoofing and Countermeasures Challenge 2017. In this study we focused on comparison of different spoofing detection approaches. These were GMM based methods, high level features extraction with simple classifier and deep learning frameworks. Experiments performed on the development and evaluation parts of the challenge dataset demonstrated stable efficiency of deep learning approac… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
3
0

Year Published

2019
2019
2022
2022

Publication Types

Select...
4
3
1

Relationship

0
8

Authors

Journals

citations
Cited by 9 publications
(3 citation statements)
references
References 21 publications
0
3
0
Order By: Relevance
“…It is noteworthy that human language entails ambiguous words, contractions, similes, jargon, and others; therefore, it takes implausible assessment and couple of minutes for blockchain and NLP to decipher the accurate output [12] and [13]. Concisely, when the needed signal has been deciphered and cognized, an acoustic echo cancellation comes in place to negate noise from the receiver signal to ensure that only the needed and intended signal remains in the system [6], [14], and [15].…”
Section: Theoretical Frameworkmentioning
confidence: 99%
“…It is noteworthy that human language entails ambiguous words, contractions, similes, jargon, and others; therefore, it takes implausible assessment and couple of minutes for blockchain and NLP to decipher the accurate output [12] and [13]. Concisely, when the needed signal has been deciphered and cognized, an acoustic echo cancellation comes in place to negate noise from the receiver signal to ensure that only the needed and intended signal remains in the system [6], [14], and [15].…”
Section: Theoretical Frameworkmentioning
confidence: 99%
“…Lavrentyeva et al [31] used a reduced version of Light Convolutional Neural Network architecture (LCNN) based on the method of the Max-Feature-Map activation (MFM) to investigate different countermeasures to defend against voice replay attacks. The LCNN with Fast Fourier-based features acquired an equal error rate of 7.34% on ASV spoof 2017 dataset in comparison to the spoofing detection method used in [32] which had an error rate of 30.74%.…”
Section: Replay Attacksmentioning
confidence: 99%
“…Categorization of countermeasures found in related studies. Lavrentyeva et al[75] explore different countermeasures to defend against voice replay attacks. Even though the countermeasure is implemented at #3 of the architecture because it needs extensive computational power, it aims at securing #1.…”
mentioning
confidence: 99%