Deep convolutional neural networks for double compressed AMR audio detection
Abstract Detection of double compressed (DC) adaptive multi‐rate (AMR) audio recordings is a challenging audio forensic problem and has received great attention in recent years. Here, the authors propose to use convolutional neural networks (CNN) for DC AMR audio detection. The CNN is used as (i) an...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-06-01
|
| Series: | IET Signal Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/sil2.12028 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832546733317423104 |
|---|---|
| author | Aykut Büker Cemal Hanilçi |
| author_facet | Aykut Büker Cemal Hanilçi |
| author_sort | Aykut Büker |
| collection | DOAJ |
| description | Abstract Detection of double compressed (DC) adaptive multi‐rate (AMR) audio recordings is a challenging audio forensic problem and has received great attention in recent years. Here, the authors propose to use convolutional neural networks (CNN) for DC AMR audio detection. The CNN is used as (i) an end‐to‐end DC AMR audio detection system and (ii) a feature extractor. The end‐to‐end system receives the audio spectrogram as the input and returns the decision whether the input audio is single compressed (SC) or DC. As a feature extractor in turn, it is used to extract discriminative features and then these features are modelled using support vector machines (SVM) classifier. Our extensive analysis conducted on four different datasets shows the success of the proposed system and provides new findings related to the problem. Firstly, double compression has a considerable impact on the high frequency components of the signal. Secondly, the proposed system yields great performance independent of the recording device or environment. Thirdly, when previously altered files are used in the experiments, 97.41% detection rate is obtained with the CNN system. Finally, the cross‐dataset evaluation experiments show that the proposed system is very effective in case of a mismatch between training and test datasets. |
| format | Article |
| id | doaj-art-5f66e389d28848929463f6db2707c3a5 |
| institution | Kabale University |
| issn | 1751-9675 1751-9683 |
| language | English |
| publishDate | 2021-06-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Signal Processing |
| spelling | doaj-art-5f66e389d28848929463f6db2707c3a52025-02-03T06:47:26ZengWileyIET Signal Processing1751-96751751-96832021-06-0115426528010.1049/sil2.12028Deep convolutional neural networks for double compressed AMR audio detectionAykut Büker0Cemal Hanilçi1Department of Electrical and Electronics Engineering Bursa Technical University Bursa TurkeyDepartment of Electrical and Electronics Engineering Bursa Technical University Bursa TurkeyAbstract Detection of double compressed (DC) adaptive multi‐rate (AMR) audio recordings is a challenging audio forensic problem and has received great attention in recent years. Here, the authors propose to use convolutional neural networks (CNN) for DC AMR audio detection. The CNN is used as (i) an end‐to‐end DC AMR audio detection system and (ii) a feature extractor. The end‐to‐end system receives the audio spectrogram as the input and returns the decision whether the input audio is single compressed (SC) or DC. As a feature extractor in turn, it is used to extract discriminative features and then these features are modelled using support vector machines (SVM) classifier. Our extensive analysis conducted on four different datasets shows the success of the proposed system and provides new findings related to the problem. Firstly, double compression has a considerable impact on the high frequency components of the signal. Secondly, the proposed system yields great performance independent of the recording device or environment. Thirdly, when previously altered files are used in the experiments, 97.41% detection rate is obtained with the CNN system. Finally, the cross‐dataset evaluation experiments show that the proposed system is very effective in case of a mismatch between training and test datasets.https://doi.org/10.1049/sil2.12028audio codingaudio recordingaudio signal processingdata compressionfeature extractionsignal classification |
| spellingShingle | Aykut Büker Cemal Hanilçi Deep convolutional neural networks for double compressed AMR audio detection IET Signal Processing audio coding audio recording audio signal processing data compression feature extraction signal classification |
| title | Deep convolutional neural networks for double compressed AMR audio detection |
| title_full | Deep convolutional neural networks for double compressed AMR audio detection |
| title_fullStr | Deep convolutional neural networks for double compressed AMR audio detection |
| title_full_unstemmed | Deep convolutional neural networks for double compressed AMR audio detection |
| title_short | Deep convolutional neural networks for double compressed AMR audio detection |
| title_sort | deep convolutional neural networks for double compressed amr audio detection |
| topic | audio coding audio recording audio signal processing data compression feature extraction signal classification |
| url | https://doi.org/10.1049/sil2.12028 |
| work_keys_str_mv | AT aykutbuker deepconvolutionalneuralnetworksfordoublecompressedamraudiodetection AT cemalhanilci deepconvolutionalneuralnetworksfordoublecompressedamraudiodetection |