Modified Group Delay Cepstral Coefficients for Voice Liveness Detection

Singh, Shrishti; Khoria, Kuldeep; Patil, Hemant A.

doi:10.23919/eusipco54536.2021.9616343

Cited by 4 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They also examined the effectiveness of different spectrum smoothing techniques in improving spectrum representation 13 . In addition, they applied the modified group delay function-based cepstral coefficients (MGDCC) as a feature to explore the effectiveness of the Gaussian Mixture Model (GMM) and the Convolutional Neaural Network (CNN) in VLD 14 . In summary, collecting PN from field microphones and further extracting spectral features is an effective strategy for VLD in the IoT.…”

Section: Related Workmentioning

confidence: 99%

FL-VLD: A Novel Voice Liveness Detection Framework Based on Asynchronous Federated Learning

Zong,

Li,

Zhu

et al. 2023

Preprint

View full text Add to dashboard Cite

Voice Liveness Detection (VLD) has become one of the hot research topics in the Internet of Things era. Conventional VLD methods are centralized solutions that are trained on abundant data collected from local clients and stored on the server, however, they have the risk of causing data islands and privacy leakage. To address this problem, we propose a novel word-level VLD framework based on asynchronous federated learning (FL) with pop noise, named FL-VLD. Structurally, FL-VLD takes the preprocessed voice for local model training and constructs a global model by only transmitting the learned weights after differential privacy with FL’s central server in an asynchronous manner. In addition, the local network of the framework incorporates the residual network and the spatial grouping enhancement module to optimize the complexity and accuracy of the global model. With the advantage of FL’s distributed structure, FL-VLD solves the data island problem in VLD scenario without threatening users’ privacy. Experimental results on the popular POCO dataset show that our proposal is clearly superior to the traditional centralized methods as well as overperforming other federated schemes in terms of fairness, stability, accuracy, and lightweightness. Further, for attacks involving far-field replay, synthesis, and conversion, FL-VLD has high generalisation capabilities. Finally, the ablation study attests to its efficacy.

show abstract

Section: Related Workmentioning

confidence: 99%