Replay attack detection method based on distortion features of speech signals introduced by loudspeaker

Abstract

The invention discloses a replay attack detection method based on the distortion features of speech signals introduced by a loudspeaker. The method includes the following steps that: speech signals to be detected are pre-processed, and noised speech frames in the speech signals to be detected are reserved; feature extraction is carried out on each noised speech frame in the pre-processed speech signals, so that speech signal linear distortion and nonlinear distortion feature-based feature vectors can be obtained; the average value of the feature vectors of all the noised speech frames is obtained, and statistical feature vectors can be obtained, and the feature model of the speech signals to be detected is obtained; the feature vectors of a training speech sample are extracted, so that a training speech feature model can be obtained, the training speech feature model is utilized to train an SVM (Support Vector Machine model), so that a speech model library can be obtained; and SVM mode matching is performed on the feature model of the speech signals to be detected and the trained speech model library, and a decision result can be outputted. With the replay attack detection method of the invention adopted, real-time and effective detection of replayed speech can be realized.

Description

technical field [0001] The invention belongs to the field of digital media processing, and relates to a playback attack detection method, in particular to a method for judging whether a voice is a voice content safe for a playback attack. Background technique [0002] The advantage of biological characteristics as an inherent attribute of living organisms has led to the emergence of biometric technology. Speaker recognition is part of biometrics, which is based on the user's voice samples to achieve identity authentication. Compared with other biological characteristics, voice has the advantages of simple voice pickup equipment, usable anytime and anywhere, and small data volume. Voiceprint verification technology has been developed for more than 60 years since it was proposed, and has made great progress, and has been widely used. However, the current identity authentication system for voiceprints is facing various masquerade attacks, including recording playback, speech sy...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the security loophole that the existing voiceprint recognition system cannot resist the playback attack, the present invention provides a playback attack detection method based on the distortion characteristics of the voice signal introduced by the loudspeaker

Images