Search

Abstract Shouted and normal speech classification plays an important role in many speech-related applications. The existing works are often based on magnitude-based features and ignore phase-based features, which are directly related to magnitude information. In this paper, the importance of phase-based features is explored for the detection of shouted speech. The novel contributions of this work are as follows. (1) Three phase-based features, namely, relative phase (RP), linear prediction analysis estimated speech-based RP (LPAES-RP) and linear prediction residual-based RP (LPR-RP) features, are explored for shouted and normal speech classification. (2) We propose a new RP feature, called the glottal source-based RP (GRP) feature. The main idea of the proposed GRP feature is to exploit the difference between RP and LPAES-RP features to detect shouted speech. (3) A score combination of phase- and magnitude-based features is also employed to further improve the classification performance. The proposed feature and combination are evaluated using the shouted normal electroglottograph speech (SNE-Speech) corpus. The experimental findings show that the RP, LPAES-RP, and LPR-RP features provide promising results for the detection of shouted speech. We also find that the proposed GRP feature can provide better results than those of the standard mel-frequency cepstral coefficient (MFCC) feature. Moreover, compared to using individual features, the score combination of the MFCC and RP/LPAES-RP/LPR-RP/GRP features yields an improved detection performance. Performance analysis under noisy environments shows that the score combination of the MFCC and the RP/LPAES-RP/LPR-RP features gives more robust classification. These outcomes show the importance of RP features in distinguishing shouted speech from normal speech.

Abstract There are many studies on detecting human speech from artificially generated speech and automatic speaker verification (ASV) that aim to detect and identify whether the given speech belongs to a given speaker. Recent studies demonstrate the success of the relative phase (RP) feature in speaker recognition/verification and the detection of synthesized speech and converted speech. However, there are few studies that focus on the RP feature for replay attack detection. In this paper, we improve the discriminating ability of the RP feature by proposing two new auditory filter-based RP features for replay attack detection. The key idea is to integrate the advantage of RP-based features in signal representation with the advantage of two auditory filter-based RP features. For the first proposed feature, we apply a Mel-filter bank to convert the signal representation of conventional RP information from a linear scale to a Mel scale, where the modified representation is called the Mel-scale RP feature. For the other proposed feature, a gammatone filter bank is applied to scale the RP information, where the scaled RP feature is called the gammatone-scale RP feature. These two proposed phase-based features are implemented to achieve better performance than a conventional RP feature because of the scale resolution and. In addition to the use of individual Mel/gammatone-scale RP features, a combination of the scores of these proposed RP features and a standard magnitude-based feature, that is, the constant Q transform cepstral coefficient (CQCC), is also applied to further improve the reliable detection decision. The effectiveness of the proposed Mel-scale RP feature, gammatone-scale RP feature, and their combination are evaluated using the ASVspoof 2017 dataset. On the evaluation dataset, our proposed methods demonstrate significant improvement over the existing feature and baseline CQCC feature. The combination of the CQCC and gammatone-scale RP provides the best performance compared with an individual baseline feature and other combination methods.

Recent studies have reported the success of linear prediction analysis (LPA)-related features, which are extracted as a short-term spectral feature for replay attack detection due to the advantage of the imperfection in the LPA-based signal produced by recording and playback devices. However, exploiting LPA-based signals is focused on only magnitude-based features and ignores phase-based features. In this paper, we propose two novel LPA-based relative phase features, namely, linear prediction residual-based relative phase (LPR-RP) and linear prediction analysis estimated speech-based relative phase (LPAES-RP). The key idea of both LPR-RP and LPAES-RP is to extract the phase information based on LPA-based signals. In the LPR-RP feature, we modify the relative phase (RP) feature extraction using a linear prediction residual (LPR) derived from the difference between the original/raw speech and LPA estimated speech signal (LPAES) instead of the original/raw speech signal. LPES-RP feature exploits the LPAES signal to replace the original/raw speech signal. Because the trace of the recording and playback device artifacts is the primary evidence for detecting the replayed signal, the advantages of the imperfection of LPR and LPAES are expected to provide efficient phase information for the replay attack detection task. In addition, using the individual LPR-RP/LPAES-RP feature, our proposed features are combined with two standard features, mel-frequency cepstral coefficients (MFCC), constant Q transform cepstral coefficients (CQCC) and the original RP feature, at score level to further improve the detection decision. The performance of the proposed LPR-RP/LPAES-RP feature and combination are evaluated using the ASVspoof 2017 version 2 database. On the evaluation subset, our proposed LPR-RP and LPAES-RP feature achieves a promising improvement over baseline features (MFCC/CQCC). Moreover, the combined systems of LPR-RP, RP, and CQCC obtains an equal error rate of 9.26%.