Yang, J., You, C., He, Q. (2018) Feature with Complementarity of Statistics and Principal Information for Spoofing Detection. Proc. Interspeech 2018, 651-655, DOI: 10.21437/Interspeech.2018-1693.
Abstract:
Constant-Q transform (CQT) has demonstrated its effectiveness in anti-spoofing feature analysis for automatic speaker verification. This paper introduces a statistics-plus-principal information feature where a short-term spectral statistics information (STSSI), octave-band principal information (OPI) and full-band principal information (FPI) are proposed on the basis of CQT. Firstly, in contrast to conventional utterance-level long-term statistic information, STSSI reveals the spectral statistics at frame-level, moreover it provides a feasibility condition for model training while only small training database is available. Secondly, OPI emphasizes the principal information for octave-bands, STSSI and OPI creates a strong complementarity to enhance the anti-spoofing feature. Thirdly, FPI is also of complementary effect with OPI. With the statistical property over CQT spectral domain and the principal information through discrete cosine transform (DCT), the proposed statistics-plus-principal feature shows reasonable advantage of the complementary trait for spoofing detection. In this paper, we setup deep neural network (DNN) classifiers for evaluation of the features. Experiments show the effectiveness of the proposed feature as compared to many conventional features on ASVspoof 2017 and ASVspoof 2015 corpus.