Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech
Voice conversion - the methodology of automatically converting one's utterances to sound as if spoken by another speaker - presents a threat for applications relying on speaker verification. We study vulnerability of text-independent speaker verification systems against voice conversion attacks...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/98757 http://hdl.handle.net/10220/13414 |
| _version_ | 1826110100480196608 |
|---|---|
| author | Kinnunen, Tomi Wu, Zhizheng Lee, Kong Aik Sedlak, Filip Chng, Eng Siong Li, Haizhou |
| author2 | School of Computer Engineering |
| author_facet | School of Computer Engineering Kinnunen, Tomi Wu, Zhizheng Lee, Kong Aik Sedlak, Filip Chng, Eng Siong Li, Haizhou |
| author_sort | Kinnunen, Tomi |
| collection | NTU |
| description | Voice conversion - the methodology of automatically converting one's utterances to sound as if spoken by another speaker - presents a threat for applications relying on speaker verification. We study vulnerability of text-independent speaker verification systems against voice conversion attacks using telephone speech. We implemented a voice conversion systems with two types of features and nonparallel frame alignment methods and five speaker verification systems ranging from simple Gaussian mixture models (GMMs) to state-of-the-art joint factor analysis (JFA) recognizer. Experiments on a subset of NIST 2006 SRE corpus indicate that the JFA method is most resilient against conversion attacks. But even it experiences more than 5-fold increase in the false acceptance rate from 3.24 % to 17.33 %. |
| first_indexed | 2024-10-01T02:29:12Z |
| format | Conference Paper |
| id | ntu-10356/98757 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:29:12Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | ntu-10356/987572020-05-28T07:41:42Z Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech Kinnunen, Tomi Wu, Zhizheng Lee, Kong Aik Sedlak, Filip Chng, Eng Siong Li, Haizhou School of Computer Engineering IEEE International Conference on Acoustics, Speech and Signal Processing (2012 : Kyoto, Japan) DRNTU::Engineering::Computer science and engineering Voice conversion - the methodology of automatically converting one's utterances to sound as if spoken by another speaker - presents a threat for applications relying on speaker verification. We study vulnerability of text-independent speaker verification systems against voice conversion attacks using telephone speech. We implemented a voice conversion systems with two types of features and nonparallel frame alignment methods and five speaker verification systems ranging from simple Gaussian mixture models (GMMs) to state-of-the-art joint factor analysis (JFA) recognizer. Experiments on a subset of NIST 2006 SRE corpus indicate that the JFA method is most resilient against conversion attacks. But even it experiences more than 5-fold increase in the false acceptance rate from 3.24 % to 17.33 %. 2013-09-09T07:33:32Z 2019-12-06T19:59:22Z 2013-09-09T07:33:32Z 2019-12-06T19:59:22Z 2012 2012 Conference Paper https://hdl.handle.net/10356/98757 http://hdl.handle.net/10220/13414 10.1109/ICASSP.2012.6288895 en © 2012 IEEE |
| spellingShingle | DRNTU::Engineering::Computer science and engineering Kinnunen, Tomi Wu, Zhizheng Lee, Kong Aik Sedlak, Filip Chng, Eng Siong Li, Haizhou Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title | Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title_full | Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title_fullStr | Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title_full_unstemmed | Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title_short | Vulnerability of speaker verification systems against voice conversion spoofing attacks : the case of telephone speech |
| title_sort | vulnerability of speaker verification systems against voice conversion spoofing attacks the case of telephone speech |
| topic | DRNTU::Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/98757 http://hdl.handle.net/10220/13414 |
| work_keys_str_mv | AT kinnunentomi vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech AT wuzhizheng vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech AT leekongaik vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech AT sedlakfilip vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech AT chngengsiong vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech AT lihaizhou vulnerabilityofspeakerverificationsystemsagainstvoiceconversionspoofingattacksthecaseoftelephonespeech |