Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network
Pitch shifting is a common voice editing technique in which the original pitch of a digital voice is raised or lowered. It is likely to be abused by the malicious attacker to conceal his/her true identity. Existing forensic detection methods are no longer effective for weakly pitch-shifted voice. In...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | International Journal of Digital Multimedia Broadcasting |
| Online Access: | http://dx.doi.org/10.1155/2020/8927031 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832547714171142144 |
|---|---|
| author | Yongchao Ye Lingjie Lao Diqun Yan Rangding Wang |
| author_facet | Yongchao Ye Lingjie Lao Diqun Yan Rangding Wang |
| author_sort | Yongchao Ye |
| collection | DOAJ |
| description | Pitch shifting is a common voice editing technique in which the original pitch of a digital voice is raised or lowered. It is likely to be abused by the malicious attacker to conceal his/her true identity. Existing forensic detection methods are no longer effective for weakly pitch-shifted voice. In this paper, we proposed a convolutional neural network (CNN) to detect not only strongly pitch-shifted voice but also weakly pitch-shifted voice of which the shifting factor is less than ±4 semitones. Specifically, linear frequency cepstral coefficients (LFCC) computed from power spectrums are considered and their dynamic coefficients are extracted as the discriminative features. And the CNN model is carefully designed with particular attention to the input feature map, the activation function and the network topology. We evaluated the algorithm on voices from two datasets with three pitch shifting software. Extensive results show that the algorithm achieves high detection rates for both binary and multiple classifications. |
| format | Article |
| id | doaj-art-7fda3998acb147fc84c9008713a0f81c |
| institution | Kabale University |
| issn | 1687-7578 1687-7586 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | International Journal of Digital Multimedia Broadcasting |
| spelling | doaj-art-7fda3998acb147fc84c9008713a0f81c2025-02-03T06:43:45ZengWileyInternational Journal of Digital Multimedia Broadcasting1687-75781687-75862020-01-01202010.1155/2020/89270318927031Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural NetworkYongchao Ye0Lingjie Lao1Diqun Yan2Rangding Wang3Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo 315211, ChinaFaculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo 315211, ChinaFaculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo 315211, ChinaFaculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo 315211, ChinaPitch shifting is a common voice editing technique in which the original pitch of a digital voice is raised or lowered. It is likely to be abused by the malicious attacker to conceal his/her true identity. Existing forensic detection methods are no longer effective for weakly pitch-shifted voice. In this paper, we proposed a convolutional neural network (CNN) to detect not only strongly pitch-shifted voice but also weakly pitch-shifted voice of which the shifting factor is less than ±4 semitones. Specifically, linear frequency cepstral coefficients (LFCC) computed from power spectrums are considered and their dynamic coefficients are extracted as the discriminative features. And the CNN model is carefully designed with particular attention to the input feature map, the activation function and the network topology. We evaluated the algorithm on voices from two datasets with three pitch shifting software. Extensive results show that the algorithm achieves high detection rates for both binary and multiple classifications.http://dx.doi.org/10.1155/2020/8927031 |
| spellingShingle | Yongchao Ye Lingjie Lao Diqun Yan Rangding Wang Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network International Journal of Digital Multimedia Broadcasting |
| title | Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network |
| title_full | Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network |
| title_fullStr | Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network |
| title_full_unstemmed | Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network |
| title_short | Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network |
| title_sort | identification of weakly pitch shifted voice based on convolutional neural network |
| url | http://dx.doi.org/10.1155/2020/8927031 |
| work_keys_str_mv | AT yongchaoye identificationofweaklypitchshiftedvoicebasedonconvolutionalneuralnetwork AT lingjielao identificationofweaklypitchshiftedvoicebasedonconvolutionalneuralnetwork AT diqunyan identificationofweaklypitchshiftedvoicebasedonconvolutionalneuralnetwork AT rangdingwang identificationofweaklypitchshiftedvoicebasedonconvolutionalneuralnetwork |