A method of synthetic speech spoofing detection using constant Q modulation envelope
In response to the low accuracy of synthetic speech spoofing detection based on traditional acoustic feature parameters, poor detection performance for unknown types of synthetic speech, and performance degradation in noisy environments, a method for detecting spoofing synthetic speech was proposed...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Beijing Xintong Media Co., Ltd
2023-11-01
|
| Series: | Dianxin kexue |
| Subjects: | |
| Online Access: | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023187/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841530889102688256 |
|---|---|
| author | Jia XU Zhihua JIAN Honghui JIN Chao WU |
| author_facet | Jia XU Zhihua JIAN Honghui JIN Chao WU |
| author_sort | Jia XU |
| collection | DOAJ |
| description | In response to the low accuracy of synthetic speech spoofing detection based on traditional acoustic feature parameters, poor detection performance for unknown types of synthetic speech, and performance degradation in noisy environments, a method for detecting spoofing synthetic speech was proposed using constant Q modulation envelope (CQME) .The motivation of the method was from the fact that the temporal envelope of speech contained abundant information and there was a big difference in detail between the envelope of synthetic speech and genuine speech.The modulation envelope spectrum of speech was obtained by employing constant Q transform (CQT), and the root mean square of each frequency component was calculated to derive the CQME feature vector.And then the CQME feature vector was used to train the random forest classifier for discriminating genuine speech from spoofing synthetic speech.Experimental results demonstrate that the random forest trained with CQME features achieves high detection performance on the ASVspoof 2019 dataset and exhibites good detection efficacy for unknown types of synthetic speech.Furthermore, the proposed method shows high detection performance even under various noise conditions, having excellent noise robustness. |
| format | Article |
| id | doaj-art-14b3b1db7f894e29ad7782d7fad53920 |
| institution | Kabale University |
| issn | 1000-0801 |
| language | zho |
| publishDate | 2023-11-01 |
| publisher | Beijing Xintong Media Co., Ltd |
| record_format | Article |
| series | Dianxin kexue |
| spelling | doaj-art-14b3b1db7f894e29ad7782d7fad539202025-01-15T02:57:57ZzhoBeijing Xintong Media Co., LtdDianxin kexue1000-08012023-11-013910711559559389A method of synthetic speech spoofing detection using constant Q modulation envelopeJia XUZhihua JIANHonghui JINChao WUIn response to the low accuracy of synthetic speech spoofing detection based on traditional acoustic feature parameters, poor detection performance for unknown types of synthetic speech, and performance degradation in noisy environments, a method for detecting spoofing synthetic speech was proposed using constant Q modulation envelope (CQME) .The motivation of the method was from the fact that the temporal envelope of speech contained abundant information and there was a big difference in detail between the envelope of synthetic speech and genuine speech.The modulation envelope spectrum of speech was obtained by employing constant Q transform (CQT), and the root mean square of each frequency component was calculated to derive the CQME feature vector.And then the CQME feature vector was used to train the random forest classifier for discriminating genuine speech from spoofing synthetic speech.Experimental results demonstrate that the random forest trained with CQME features achieves high detection performance on the ASVspoof 2019 dataset and exhibites good detection efficacy for unknown types of synthetic speech.Furthermore, the proposed method shows high detection performance even under various noise conditions, having excellent noise robustness.http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023187/synthetic speechspoofing speech detectionconstant Q modulation enveloperandom forest |
| spellingShingle | Jia XU Zhihua JIAN Honghui JIN Chao WU A method of synthetic speech spoofing detection using constant Q modulation envelope Dianxin kexue synthetic speech spoofing speech detection constant Q modulation envelope random forest |
| title | A method of synthetic speech spoofing detection using constant Q modulation envelope |
| title_full | A method of synthetic speech spoofing detection using constant Q modulation envelope |
| title_fullStr | A method of synthetic speech spoofing detection using constant Q modulation envelope |
| title_full_unstemmed | A method of synthetic speech spoofing detection using constant Q modulation envelope |
| title_short | A method of synthetic speech spoofing detection using constant Q modulation envelope |
| title_sort | method of synthetic speech spoofing detection using constant q modulation envelope |
| topic | synthetic speech spoofing speech detection constant Q modulation envelope random forest |
| url | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2023187/ |
| work_keys_str_mv | AT jiaxu amethodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT zhihuajian amethodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT honghuijin amethodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT chaowu amethodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT jiaxu methodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT zhihuajian methodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT honghuijin methodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope AT chaowu methodofsyntheticspeechspoofingdetectionusingconstantqmodulationenvelope |