FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning
Deep learning brings the opportunity to achieve effective speech steganalysis in speech signals. However, the speech samples used to train speech steganalysis models (i.e., steganalyzers) are usually sensitive and distributed among different agencies, making it impractical to train an effective cent...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-06-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/12/13/2854 |
_version_ | 1797591819772493824 |
---|---|
author | Hui Tian Huidong Wang Hanyu Quan Wojciech Mazurczyk Chin-Chen Chang |
author_facet | Hui Tian Huidong Wang Hanyu Quan Wojciech Mazurczyk Chin-Chen Chang |
author_sort | Hui Tian |
collection | DOAJ |
description | Deep learning brings the opportunity to achieve effective speech steganalysis in speech signals. However, the speech samples used to train speech steganalysis models (i.e., steganalyzers) are usually sensitive and distributed among different agencies, making it impractical to train an effective centralized steganalyzer. Therefore, in this paper, we present an effective framework, named FedSpy, using federated learning, which enables multiple agencies to securely and jointly train the speech steganalysis models without sharing their speech samples. FedSpy is a flexible and extensible framework that can work effectively in conjunction with various deep-learning-based speech steganalysis methods. We evaluate the performance of FedSpy by detecting the most widely used Quantization Index Modulation-based speech steganography with three state-of-the-art deep-learning-based steganalysis methods representatively. The results show that FedSpy significantly outperforms the local steganalyzers and achieves good detection accuracy comparable to the centralized steganalyzer. |
first_indexed | 2024-03-11T01:43:48Z |
format | Article |
id | doaj.art-35422b8372234eaf8f9b32155eba964e |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-11T01:43:48Z |
publishDate | 2023-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-35422b8372234eaf8f9b32155eba964e2023-11-18T16:24:23ZengMDPI AGElectronics2079-92922023-06-011213285410.3390/electronics12132854FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated LearningHui Tian0Huidong Wang1Hanyu Quan2Wojciech Mazurczyk3Chin-Chen Chang4College of Computer Science and Technology, National Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, National Huaqiao University, Xiamen 361021, ChinaCollege of Computer Science and Technology, National Huaqiao University, Xiamen 361021, ChinaInstitute of Computer Science, Warsaw University of Technology, 00-665 Warszawa, PolandDepartment of Information and Computer Science, Feng Chia University, Taichung 40724, TaiwanDeep learning brings the opportunity to achieve effective speech steganalysis in speech signals. However, the speech samples used to train speech steganalysis models (i.e., steganalyzers) are usually sensitive and distributed among different agencies, making it impractical to train an effective centralized steganalyzer. Therefore, in this paper, we present an effective framework, named FedSpy, using federated learning, which enables multiple agencies to securely and jointly train the speech steganalysis models without sharing their speech samples. FedSpy is a flexible and extensible framework that can work effectively in conjunction with various deep-learning-based speech steganalysis methods. We evaluate the performance of FedSpy by detecting the most widely used Quantization Index Modulation-based speech steganography with three state-of-the-art deep-learning-based steganalysis methods representatively. The results show that FedSpy significantly outperforms the local steganalyzers and achieves good detection accuracy comparable to the centralized steganalyzer.https://www.mdpi.com/2079-9292/12/13/2854speech steganalysisspeech steganographyfederated learning |
spellingShingle | Hui Tian Huidong Wang Hanyu Quan Wojciech Mazurczyk Chin-Chen Chang FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning Electronics speech steganalysis speech steganography federated learning |
title | FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning |
title_full | FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning |
title_fullStr | FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning |
title_full_unstemmed | FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning |
title_short | FedSpy: A Secure Collaborative Speech Steganalysis Framework Based on Federated Learning |
title_sort | fedspy a secure collaborative speech steganalysis framework based on federated learning |
topic | speech steganalysis speech steganography federated learning |
url | https://www.mdpi.com/2079-9292/12/13/2854 |
work_keys_str_mv | AT huitian fedspyasecurecollaborativespeechsteganalysisframeworkbasedonfederatedlearning AT huidongwang fedspyasecurecollaborativespeechsteganalysisframeworkbasedonfederatedlearning AT hanyuquan fedspyasecurecollaborativespeechsteganalysisframeworkbasedonfederatedlearning AT wojciechmazurczyk fedspyasecurecollaborativespeechsteganalysisframeworkbasedonfederatedlearning AT chinchenchang fedspyasecurecollaborativespeechsteganalysisframeworkbasedonfederatedlearning |