Gauri P. Prajapati's research while affiliated with Dhirubhai Ambani Institute of Information and Communication Technology and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (6)
Attempts to spoof an ASV (voice biometric system) have been successful in the past due to the advent of technologies. However, despite the development of various countermeasures for each spoofing attack, there is an urgent need for a versatile countermeasure. Hence, designing a voice privacy system has become crucial. Moreover, the energy losses in...
Extensive use of Intelligent Personal Assistants (IPA) and biometrics in our day-to-day life asks for privacy preservation while dealing with personal data. To that effect, efforts have been made to preserve the personally identifiable characteristics from human voice using different speaker anonymization techniques. In this paper, we propose Cycle...
Speaker's identity is the most crucial information exploited (implicitly) by an Automatic Speaker Verification (ASV) system. Numerous attacks can be obliterated simultaneously if privacy preservation is exercised for a speaker's identity. The baseline of the Voice Privacy Challenge 2020 by INTER-SPEECH uses the Linear Prediction (LP) model of speec...
Citations
... Specifically, they altered the pole angles of the linear prediction (LP) spectral envelope via the McAdams coefficient [51]. Gupta et al. [29] improved this work by modifying both the pole angles and the pole radii of the LP spectral envelope. Although these parameter manipulations are perceptually reasonable, speaker information can be easily recovered by machine learning-based attacks, such as training an ASI or ASV system on transformed speech [71]. ...
Reference: Differentially Private Speaker Anonymization
![[object Object]](https://i1.rgstatic.net/ii/profile.image/922561227534336-1596967145537_Q64/Priyanka-Gupta-5.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/787886123253760-1564858098336_Q64/Madhu-Kamble.jpg)
... Trust in voice assistant technology and service providers, is fundamental, especially for non-users [2,3,4]. Recently, many studies on voice anonymization have been initiated (e.g., Prajapati et al. [5]). Especially regarding increasing user acceptance and trust, anonymization represents a promising approach. ...
... Several speaker anonymization systems use adversarial autoencoders or GANs to improve voice privacy. However, they use them to either disentangle multiple private attributes like sex and accent in addition to identity in order to increase control in anonymization [18,28] -and even to change an attribute like sex but keeping the original identity [26] -, or to disentangle the speaker information from the speech without using explicit speaker embeddings [24]. Although GANs have been applied to speaker embeddings for data augmentation [29], to our knowledge, no work has yet attempted to use this for speaker anonymization by mimicking the properties of the original speaker vector space in order to sample artificial yet natural-like embeddings. ...
... To detect the first-and second-order replay attacks, Malik et al. developed an antispoofing framework combining the acoustic ternary patterns Gammatone cepstral coefficient features and multiclass SVM classifier [32]. Prajapati et al. designed a replay voice detection system for voice assistants using the energy separation method and the Gaussian mixture model (GMM) classifier [33]. Furthermore,in this application the combination of Teager energy cepstrum coefficients (TECC) feature and classifier (GMM or light CNN) was considered [34]. ...
