Fusion of biLSTM and GMM-UBM systems for audio spoofing detection

Статья в журнале

In this paper, we present our contribution to the ASVspoof 2019 challenge. The main task for this challenge is to find countermeasures that generalize well for different spoofing attacks against automatic speaker verification systems. Some of the approaches used by the authors during participation in the challenge are presented. Described anti-spoofing systems mostly rely on using constant Q cepstral coefficients (CQCC) features and bidirectional long-short term memory (BiLSTM) networks for genuine/spoof audio classification. Fusion of BiLSTM and GMM-UBM system is presented. This approach could give significant improvement to baseline systems results without any data augmentation, especially on physical access (PA) condition. Presented systems give 15.2% min-tDCF relative improvement for logical access (LA) condition and 61.5% min-tDCF relative improvement for PA condition, compared to the best baseline systems results.

Журнал:

• International Journal of Advanced Trends in Computer Science and Engineering
• World Academy of Research in Science and Engineering (Tiruppūr)
• Индексируется в Scopus

Библиографическая запись: Rakhmanenko, I. A. Fusion of biLSTM and GMM-UBM systems for audio spoofing detection / I. A. Rakhmanenko, A. A. Shelupanov, E. Y. Kostyuchenko. // International Journal of Advanced Trends in Computer Science and Engineering. - 2019. - Vol. 8. - № 4. - P. 1741-1746. - DOI: 10.30534/ijatcse/2019/103842019

Индексируется в:

• Scopus ( https://www.scopus.com/record/display.uri?eid=2-s2.0-85073392345&origin=resultslist&sort=plf-f&src=s&sid=eb612801eac0dc5de7f68a0d772966df&sot=autdocs&sdt=autdocs&sl=18&s=AU-ID%2857190966992%29&relpos=0&citeCnt=0&searchTerm= )