A Speech Steganalysis Algorithm Based on Multi-Feature Fusion and BiLSTM

SU Zhao-pin; ZHANG Ling; ZHANG Guo-fu; YUE Feng

doi:10.12263/DZXB.20220553

您当前的位置：

首页 >

文章列表页 >

A Speech Steganalysis Algorithm Based on Multi-Feature Fusion and BiLSTM

PAPERS | 更新时间：2025-12-08

- A Speech Steganalysis Algorithm Based on Multi-Feature Fusion and BiLSTM
- ACTA ELECTRONICA SINICA Vol. 51, Issue 5, Pages: 1300-1309(2023)
- 作者机构：
  
  1.合肥工业大学计算机与信息学院, 安徽合肥 230601
  2.大数据知识工程教育部重点实验室(合肥工业大学),安徽合肥 230601
  3.智能互联系统安徽省实验室(合肥工业大学),安徽合肥 230009
  4.工业安全应急技术安徽省重点实验室(合肥工业大学),安徽合肥 230601
- 作者简介：
- 基金信息：
  
  Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation(2020B121201001)
- DOI：10.12263/DZXB.20220553
  CLC： TP309
- Received：14 May 2022，
  
  Revised：2022-06-24，
  
  Published：25 May 2023
- 稿件说明：
移动端阅览
苏兆品,张羚,张国富等.基于多特征融合和BiLSTM的语音隐写检测算法[J].电子学报,2023,51(05):1300-1309.

SU Zhao-pin,ZHANG Ling,ZHANG Guo-fu,et al.A Speech Steganalysis Algorithm Based on Multi-Feature Fusion and BiLSTM[J].ACTA ELECTRONICA SINICA,2023,51(05):1300-1309.
苏兆品,张羚,张国富等.基于多特征融合和BiLSTM的语音隐写检测算法[J].电子学报,2023,51(05):1300-1309. DOI： 10.12263/DZXB.20220553.

SU Zhao-pin,ZHANG Ling,ZHANG Guo-fu,et al.A Speech Steganalysis Algorithm Based on Multi-Feature Fusion and BiLSTM[J].ACTA ELECTRONICA SINICA,2023,51(05):1300-1309. DOI： 10.12263/DZXB.20220553.

摘要

针对传统互联网低比特率编解码器（internet Low Bit Rate Codec，iLBC）语音隐写主要集中在线性频谱频率系数矢量量化、码本搜索矢量量化或增益量化的单个阶段，难以应对多阶段下的联合隐写检测等问题，提出一种基于多特征融合和双向长短时记忆（Bi-Directional Long Short-Term Memory，BiLSTM）网络的iLBC语音隐写检测算法.通过分析隐写对不同阶段参数带来的影响，提取线性频谱频率系数矢量量化、码本搜索矢量量化和增益量化过程中的多种隐写特征，并分别输入到相应的BiLSTM检测网络，最后将各检测网络的结果进行融合，得到最终隐写检测结果.实验表明，所提算法可以实现多阶段下的联合隐写检测，而且在语音时长较短时，仍能取得优异的检测结果，平均检测准确率达到了90%以上.

Abstract

The traditional internet low bit rate codec (iLBC) based speech steganography mainly focuses on a single stage of the linear spectrum frequency coefficient vector quantization

the codebook search vector quantization

or the gain quantization

which is difficult to deal with the multi-stage joint steganalysis. To this end

an iLBC speech steganalysis algorithm based on the multi-feature fusion and the bi-directional long short-term memory (BiLSTM) network is proposed. Specifically

the impact of steganography on iLBC parameters is first analyzed in the linear spectrum frequency coefficient vector quantization process

the dynamic codebook search process

and the gain quantization process. Then

multiple steganographic features in the above three stages are extracted and input to three different detection models based on BiLSTM

respectively. Finally

a fusion strategy is presented to merge the detection results of each model. Experimental results show that the proposed algorithm can achieve multi-stage joint steganalysis and good detection results with an average detection accuracy of more than 90%

even if the speech duration is short.

关键词

Keywords

references

WU Z , SHA Y . An implementation of speech steganography for iLBC by using fixed codebook [C]// IEEE International Conference on Computer and Communications . Chengdu : IEEE Press , 2016 : 1970 - 1974 .

HUANG Y , TAO H , XIAO B , et al . Steganography in low bit-rate speech streams based on quantization index modulation controlled by keys [J]. Science China Technological Sciences , 2017 , 60 ( 10 ): 1585 - 1596 .

SU Z , LI W , ZHANG G , et al . A steganographic method based on gain quantization for iLBC speech streams [J]. Multimedia Systems , 2020 , 26 ( 2 ): 223 - 233 .

苏兆品 , 张羚 , 张国富 . 低比特率语音流大容量分层隐写方法 [J/OL]. 中国图象图形学报 , 2022 , DOI: 10.11834/jig.210307 http://dx.doi.org/10.11834/jig.210307 .

SU Zhao-pin , ZHANG Ling , ZHANG Guo-fu . High-capacity hierarchical steganography in a low-bit rate speech codec [J/OL]. Journal of Image and Graphics , 2022 , DOI: 10.11834/jig.210307. http://dx.doi.org/10.11834/jig.210307. (in Chinese)

LIU Q , SUNG A H , QIAO M . Temporal derivative-based spectrum and mel-cepstrum audio steganalysis [J]. IEEE Transactions on Information Forensics and Security , 2009 , 4 ( 3 ): 359 - 368 .

LIN Z , HUANG Y , WANG J . RNN-SM: Fast steganalysis of VoIP streams using recurrent neural network [J]. IEEE Transactions on Information Forensics and Security , 2018 , 13 ( 7 ): 1854 - 1868 .

GONG C , YI X , ZHAO X , et al . Recurrent convolutional neural networks for AMR steganalysis based on pulse position [C]// ACM Workshop on Information Hiding and Multimedia Security , Paris : ACM Press , 2019 : 2 - 13 .

REN Y , LIU D , LIU C , et al . A universal audio steganalysis scheme based on multiscale spectrograms and deep-ResNet [J/OL]. IEEE Transactions on Dependable and Secure Computing , 2022 . DOI: 10.1109/TDSC.2022.3141121 http://dx.doi.org/10.1109/TDSC.2022.3141121 .

YANG H , YANG Z , BAO Y , et al . FCEM: A novel fast correlation extract model for real time steganalysis of VoIP stream via multi-head attention [C]// International Conference on Acoustics, Speech and Signal Processing , Barcelona : IEEE , 2020 : 2822 - 2826 .

YANG H , YANG Z , BAO Y , et al . Fast steganalysis method for VoIP streams [J]. IEEE Signal Processing Letters , 2020 , 14 : 286 - 290 .

李望望 . 面向iLBC语音流的隐写与隐写分析技术研究 [D]. 合肥 : 合肥工业大学计算机与信息学院 , 2019 .

张浩 , 胡昌华 , 杜党波等 . 多状态影响下基于Bi-LSTM网络的锂电池剩余寿命预测方法 [J]. 电子学报 , 2022 , 50 ( 3 ): 619 - 624 .

ZHANG H , HU C , DU D , et al . Remaining useful life prediction method of lithium-ion battery based on Bi-LSTM network under multi-state influence [J]. Acta Electronica Sinica , 2022 , 50 ( 3 ): 619 - 624 . (in Chinese)

李敬轩 , 胡润文 , 阮观奇 , 等 . 基于手工特征提取与结果融合的CNN音频隐写分析算法 [J]. 计算机学报 , 2021 , 44 ( 10 ): 2061 - 2075 .

LI J , HU R , RUAN G , et al . A CNN based audio steganalysis algorithm by manual feature extraction and result merging [J]. Chinese Journal of Computers , 2021 , 44 ( 10 ): 2061 - 2075 . (in Chinese)

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-Static Underwater Small Target Recognition Based on Kernel Joint Sparse Representation and Exponential Smoothing

Multi-Feature Gait Pattern Recognition Based on Fisher Discriminant and GKF-RELM Algorithm

Research on Multi-feature Fusion Impact Authentication for Intelligent Mobile Device

Multi-feature Fusion Based Retrieval Results Optimization for Crime Scene Investigation Image Retrieval

Related Author

Zhao-pin SU

Ling ZHANG

Guo-fu ZHANG

Feng YUE

YANG Juan

XU Feng

WANG Jia-wei

GAO Fa-rong

Related Institution

School of Computer Science and Information Engineering， Hefei University of Technology

Key Laboratory of Knowledge Engineering with Big Data（Hefei University of Technology）， Ministry of Education

University of Chinese Academy of Sciences

Institute of Acoustics， Chinese Academy of Sciences

School of Automation, Hangzhou Dianzi University

⁰