Score normalization in playback attack detection

Abstract

The task of a playback attack detector (PAD) is to decide whether an incoming recording shares the same originating utterance as any of N stored recordings. All recordings are noisy channel-distorted versions of the same phrase uttered by the same person; the originating utterances of the N stored recordings are assumed to be distinct. The proposed approach makes a decision based on a set of N similarity scores which quantify the similarity between the incoming recording and each of the N stored recordings. Although satisfactory results are obtained by thresholding the maximum of the N scores using speaker and phrase (SaP)-dependent thresholds, it is shown that the use of a relative similarity score (a normalized version of the maximum similarity score) results in significant performance improvements especially in the case when the incoming recording is a severely distorted version of a stored recording utterance, as well as for the case when SaP-independent thresholds are used.

Speaker Recognition Anti-spoofing

“…A countermeasure for replay attack detection in the case of text-dependent ASV was reported in [36]. The approach is based upon the comparison of new access samples with stored instances of past accesses.…”

Spoofing and countermeasures for speaker verification: A survey

“…Recently, due to the mass-market adoption of ASV techniques Nuance, 2013) and the awareness and simplicity of replay attacks, both industry (Nuance, 2013) and academia (Shang and Stevenson, 2010;Villalba and Lleida, 2011a,b;Wang et al, 2011) have shown an interest in developing replay attack countermeasures.…”

“…The first approach to replay detection was reported in (Shang and Stevenson, 2010) in the context of a text-dependent ASV system using fixed pass-phrases. The detector is based upon the comparison of new access samples with stored instances of past access attempts.…”

Voice Playback Detection based on Long-window Scale-factors

“…Besides, the algorithm was improved in [5] in which the spectral peaks are combined into pairs with a seed peak. This method, in [2][3][4][5], is only based on text-dependent speaker verification system and the efficiency of detection will significantly decrease When the number of request certification for client increase. 2) Detection algorithm based on the speech channel.…”

“…In other words, though the same person says the same content at different times, the speech signals still have great randomness. In [2] and [3], a playback detector based on the random of speech sounds was designed to compare the verification recording with stored recording of past access attempts. Wu,in [4], adopted the same algorithm on the study of playback speech detection and extended the experimental samples.…”