Can We Use Speaker Recognition Technology to Attack Itself? Enhancing Mimicry Attacks Using Automatic Target Speaker Selection

Tomi Kinnunen 1 Rosa Hautamäki 1 Ville Vestman 1 Md Sahidullah 2
2 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : We consider technology-assisted mimicry attacks in the context of automatic speaker verification (ASV). We use ASV itself to select targeted speakers to be attacked by human-based mimicry. We recorded 6 naive mimics for whom we select target celebrities from VoxCeleb1 and VoxCeleb2 corpora (7,365 potential targets) using an i-vector system. The attacker attempts to mimic the selected target, with the utterances subjected to ASV tests using an independently developed x-vector system. Our main finding is negative: even if some of the attacker scores against the target speakers were slightly increased, our mimics did not succeed in spoofing the x-vector system. Interestingly, however, the relative ordering of the selected targets (closest, furthest, median) are consistent between the systems, which suggests some level of transferability between the systems.
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download

https://hal.inria.fr/hal-02051701
Contributor : Md Sahidullah <>
Submitted on : Thursday, February 28, 2019 - 4:46:54 AM
Last modification on : Wednesday, April 3, 2019 - 1:23:16 AM
Long-term archiving on: Wednesday, May 29, 2019 - 5:30:49 PM

File

ICASSP19_Manuscript_UEF_Inria....
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02051701, version 1

Collections

Citation

Tomi Kinnunen, Rosa Hautamäki, Ville Vestman, Md Sahidullah. Can We Use Speaker Recognition Technology to Attack Itself? Enhancing Mimicry Attacks Using Automatic Target Speaker Selection. ICASSP 2019 – 44th International Conference on Acoustics, Speech, and Signal Processing, May 2019, Brighton, United Kingdom. ⟨hal-02051701⟩

Share

Metrics

Record views

87

Files downloads

140