Identification of Electronic Disguised Voices in the Noisy Environment

Cao, Wencheng; Wang, Hongxia; Hong, Zhao; Qian, Qing; Abdullahi, Sani M.

doi:10.1007/978-3-319-53465-7_6

Cited by 5 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each voice sample, 20-dimensional LFCC feature map is extracted by setting the length of frame to 256 and the number of lters to 20 in Equation (2). In [6], LFCC with SVM classi er achieves great robustness detecting disguised voice in noisy environment. In our work, the GMM classi er of 64640 voice samples.…”

Section: Proposed Identi Cation Algorithm For Pitch-shi Ed Voicementioning

confidence: 99%

“…e rst row and the last row in Figure 7 indicate the intra-dataset results, the detection rates of proposed method are higher than 90% in most cases, while others are greatly a ected by di erent pitch shi ing so ware and even drop lower than 60%. e 2 nd and 3 rd rows show the cross-dataset results, especially for a few speci c semitones, both [6] and [8] lower than 20%. Proposed methods remain a steady performance with the worst case of ~60% and ~80% for most cases.…”

Section: Weakly Pitch-shi Edmentioning

confidence: 99%

“…In this case, as a comparison to [6] and [8], we focus on the voice strongly shi ed with factors from ±5 to ±8 semitones. Firstly, we try to identify whether the suspected voice is original or pitch-shi ed one.…”

Section: Strongly Pitch-shi Edmentioning

confidence: 99%

“…Especially for the voices shi ed with ±4 semitones, the detection rates drop lower than 90%. In [6], environment noise is considered in identifying the pitch shi ing.…”

Section: Introductionmentioning

confidence: 99%

“…e experimental results show that the features extracted from linear frequency cepstral coe cients (LFCC) and formant can be e ectively discriminant the natural and pitch-shi ed voice. However, the experimental results on weakly pitch-shi ed voices have not been given in [6].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Identification of Weakly Pitch-Shifted Voice Based on Convolutional Neural Network

Lao

Yan

et al. 2020

International Journal of Digital Multimedia Broadcasting

View full text Add to dashboard Cite

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.

show abstract

Section: Proposed Identi Cation Algorithm For Pitch-shi Ed Voicementioning

confidence: 99%

Section: Weakly Pitch-shi Edmentioning

confidence: 99%

Section: Strongly Pitch-shi Edmentioning

confidence: 99%

“…Especially for the voices shi ed with ±4 semitones, the detection rates drop lower than 90%. In [6], environment noise is considered in identifying the pitch shi ing.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations