Spoofing-aware speaker verification (SASV) is an important but challenging task and has been a primary focus of the recently organized ASVspoof 5 challenge. As SASV integrates automatic speaker verification (ASV) and countermeasure (CM) systems, its performance depends on the effectiveness of each system. This study systematically examines the impact of different parameter-sharing (PS) strategies, which facilitate joint optimization, on SASV performance using the ASVspoof 5 dataset. Experimental results indicate that PS enhances performance for specific attack types and codec conditions. For example, the baseline system achieves a min a-DCF of 0.329 on the A26 attack, which improves to 0.233 with PS. Similarly, for AMR-compressed signals, PS yields a 14.09% performance gain. These observations show that PS techniques are effective in mitigating certain spoofing attacks and improving robustness to degraded audio conditions in SASV systems.

Automatic speaker verification (ASV) systems are vulnerable to spoofing attacks. We propose a spoofing-robust ASV system optimized directly for the recently introduced architecture-agnostic detection cost function (a-DCF), which allows targeting a desired trade-off between the contradicting aims of user convenience and robustness to spoofing. We combine a-DCF and binary cross-entropy (BCE) with a novel straightforward threshold optimization technique. Our results with an embedding fusion system on ASVspoof2019 data demonstrate relative improvement of 13% over a system trained using BCE only (from minimum a-DCF of 0.1445 to 0.1254). Using an alternative non-linear score fusion approach provides relative improvement of 43% (from minimum a-DCF of 0.0508 to 0.0289).

Double compressed (DC) adaptive multi-rate (AMR) audio detection is an important but challenging audio forensic task which has received great attention over the last decade. Although the majority of the existing studies extract hand-crafted features and classify these features using traditional pattern matching algorithms such as support vector machines (SVM), recently convolutional neural network (CNN) based DC AMR audio detection system was proposed which yields very promising detection performance. Similar to any traditional CNN based classification system, CNN based DC AMR recognition system uses standard softmax loss as the training criterion. In this paper, we propose to use angular margin softmax loss and its variants for DC AMR detection problem. Although using angular margin softmax was originally proposed for face recognition, we adapt it to the CNN based end-to-end DC audio detection system. The angular margin softmax basically introduces a margin between two classes so that the system can learn more discriminative embeddings for the problem. Experimental results show that adding angular margin penalty to the traditional softmax loss increases the average DC AMR audio detection from 95.83% to 100%. It is also found that the angular margin softmax loss functions boost the DC AMR audio detection performance when there is a mismatch between training and test datasets.

Abstract Detection of double compressed (DC) adaptive multi‐rate (AMR) audio recordings is a challenging audio forensic problem and has received great attention in recent years. Here, the authors propose to use convolutional neural networks (CNN) for DC AMR audio detection. The CNN is used as (i) an end‐to‐end DC AMR audio detection system and (ii) a feature extractor. The end‐to‐end system receives the audio spectrogram as the input and returns the decision whether the input audio is single compressed (SC) or DC. As a feature extractor in turn, it is used to extract discriminative features and then these features are modelled using support vector machines (SVM) classifier. Our extensive analysis conducted on four different datasets shows the success of the proposed system and provides new findings related to the problem. Firstly, double compression has a considerable impact on the high frequency components of the signal. Secondly, the proposed system yields great performance independent of the recording device or environment. Thirdly, when previously altered files are used in the experiments, 97.41% detection rate is obtained with the CNN system. Finally, the cross‐dataset evaluation experiments show that the proposed system is very effective in case of a mismatch between training and test datasets.

Automatic speaker verification (ASV) systems can be easily spoofed by previously recorded speech, synthesized speech and speech signal that artificially generated by voice conversion techniques. In order to increase the reliability of the ASV systems, detecting spoofing attacks whether a given speech signal is genuine or spoofed plays an important role. In this paper, we consider the detection of replay attacks which is the most accessible attack type against ASV systems. To this end, we utilize a deep neural network (DNN) based classifier using features extracted from the long-term average spectrum. The experiments are conducted on the latest edition of Automatic Speaker Verification Spoofing and Countermeasures Challenge (ASVspoof 2017) database. The results are compared with the ASVspoof 2017 baseline system which consists of Gaussian mixture model (GMM) classifier with constant-Q transform cepstral coefficients (CQCC) front-end as well as the GMM with standard mel-frequency cepstrum coefficients (MFCC) features. Experimental results reveal that DNN considerably outperforms the well-known and successful GMM classifier. It is found that long term average spectrum (LTAS) based features are superior to CQCC and MFCC in terms of equal error rate (EER). Finally, we find that high-frequency components convey much more discriminative information for replay attack detection independent of features and classifiers.

Automatic speaker verification (ASV) systems are highly vulnerable against spoofing attacks, also known as imposture. With recent developments in speech synthesis and voice conversion technology, it has become important to detect synthesized or voice-converted speech for the security of ASV systems. In this paper, we compare five different classifiers used in speakerrecognition to detect synthetic speech. Experimental results conducted on the ASVspoof 2015 dataset show that supportvector machines with generalized linear discriminant kernel(GLDS-SVM) yield the best performance on the development set with the EER of 0.12 % whereas Gaussian mixture model (GMM) trained using maximum likelihood (ML) criterion with the EER of 3.01 % is superior for the evaluation set.

The performance of biometric systems based on automatic speaker recognition technology is severely degraded due to spoofing attacks with synthetic speech generated using different voice conversion (VC) and speech synthesis (SS) techniques. Various countermeasures are proposed to detect this type of at-tack, and in this context, choosing an appropriate feature extrac-tion technique for capturing relevant information from speech is an important issue. This paper presents a concise experi-mental review of different features for synthetic speech detec-tion task. A wide variety of features considered in this study include previously investigated features as well as some other potentially useful features for characterizing real and synthetic speech. The experiments are conducted on recently released ASVspoof 2015 corpus containing speech data from a large number of VC and SS technique. Comparative results using two different classifiers indicate that features representing spectral information in high-frequency region, dynamic information of speech, and detailed information related to subband characteris-tics are considerably more useful in detecting synthetic speech. Index Terms: anti-spoofing, ASVspoof 2015, feature extrac-tion, countermeasures

An increasing number of independent studies have con-firmed the vulnerability of automatic speaker verification (ASV) technology to spoofing. However, in comparison to that involving other biometric modalities, spoofing and countermea-sure research for ASV is still in its infancy. A current barrier to progress is the lack of standards which impedes the comparison of results generated by different researchers. The ASVspoof ini-tiative aims to overcome this bottleneck through the provision of standard corpora, protocols and metrics to support a common evaluation. This paper introduces the first edition, summaries the results and discusses directions for future challenges and re-search.

This paper describes an approach to robust signal analysis using iterative parameter re-estimation of a mixture autoregressive (AR) model. The model's focus can be adjusted by initialization of the target and non-target states. The variant examined in this study uses an i.i.d. mixture AR model and is designed to tackle the spectral biasing effect caused by the voice excitation in speech signals with variable fundamental frequency. In our speaker verification experiments, this method performed competitively against standard spectrum analysis techniques in non-mismatch conditions and showed significant improvements in vocal effort mismatch conditions.

Text-independent speaker identification is studied using neutral and shouted speech in Finnish to analyze the effect of vocal mode mismatch between training and test utterances. Standard mel-frequency cepstral coefficient (MFCC) features with Gaussian mixture model (GMM) recognizer are used for speaker identification. The results indicate that speaker identification accuracy reduces from perfect (100 %) to 8.71 % under vocal mode mismatch. Because of this dramatic degradation in recognition accuracy, we propose to use a joint density GMM mapping technique for compensating the MFCC features. This mapping is trained on a disjoint emotional speech corpus to create a completely speaker- and speech mode independent emotion-neutralizing mapping. As a result of the compensation, the 8.71 % identification accuracy increases to 32.00 % without degrading the non-mismatched train-test conditions much.

This paper presents comparison and optimization of acoustic features for source cell-phone recognition using recorded speech signals. Different acoustic feature extraction methods such as Mel-frequency, linear frequency and Bark frequency cepstral coefficients (MFCC, LFCC and BFCC) and linear prediction cepstral coefficients (LPCC) are considered. In addition to different feature sets, the effect of dynamic features, delta and double-delta coefficients (Δ and Δ2), and feature normalizations, cepstral mean normalization (CMN), cepstral variance normalization (CVN) and cepstral mean and variance normalization (CMVN) are also examined on the performance of source cell-phone recognition. The same support vector machine (SVM) classifier with fixed parameters and the same cell-phone dataset are used in the experiments in order to make a fair comparison of different features and feature normalization techniques.

We study the problem of vocal effort mismatch in speaker verification. Changes in speaker’s vocal effort induce changes in fundamental frequency (F0) and formant structure which introduce unwanted intra-speaker variations to features. We compare seven alternative spectrum estimators in the context of melfrequency cepstral coefficient (MFCC) extraction for speaker verification. The compared variants include traditional FFT spectrum and six parametric all-pole models. Experimental results on the NIST 2010 speaker recognition evaluation (SRE) corpus utilizing both GMM-UBM and more recent GMM supervector classifier indicate that spectrum estimation has a considerable impact on speaker verification accuracy under mismatched vocal effort conditions. The highest recognition accuracy was achieved using a particular variant of temporally weighted all-pole model, stabilized weighted linear prediction (SWLP). Index Terms: speaker recognition, vocal effort mismatch, spectrum estimation

Popular features for speech processing, such as mel-frequency cepstral coefficients (MFCCs), are derived from the short-term magnitude spectrum, whereas the phase spectrum remains un-used. While the common argument to use only the magnitude spectrum is that the human ear is phase-deaf, phase-based fea-tures have remained less explored due to additional signal pro-cessing difficulties they introduce. A useful representation of the phase is the group delay function, but its robust computa-tion remains difficult. This paper advocates the use of group delay functions derived from parametric all-pole models instead of their direct computation from the discrete Fourier transform. Using a subset of the vocal effort data in the NIST 2010 speaker recognition evaluation (SRE) corpus, we show that group delay features derived via parametric all-pole models improve recog-nition accuracy, especially under high vocal effort. Addition-ally, the group delay features provide comparable or improved accuracy over conventional magnitude-based MFCC features. Thus, the use of group delay functions derived from all-pole models provide an effective way to utilize information from the phase spectrum of speech signals. Index Terms: speaker verification, group delay functions, high vocal effort

I4U is a joint entry of nine research Institutes and Universities across 4 continents to NIST SRE 2012. It started with a brief discussion during the Odyssey 2012 workshop in Singapore. An online discussion group was soon set up, providing a discussion platform for different issues surrounding NIST SRE’12. Noisy test segments, uneven multi-session training, variable enrollment duration, and the issue of open-set identification were actively discussed leading to various solutions integrated to the I4U submission. The joint submission and several of its 17 sub-systems were among top-performing systems. We summarize the lessons learnt from this large-scale effort.

Different short-term spectrum estimators for speaker verification under additive noise are considered. Conventionally, mel-frequency cepstral coefficients (MFCCs) are computed from discrete Fourier transform (DFT) spectra of windowed speech frames. Recently, linear prediction (LP) and its temporally weighted variants have been substituted as the spectrum analysis method in speech and speaker recognition. In this paper, 12 different short-term spectrum estimation methods are compared for speaker verification under additive noise contamination. Experimental results conducted on NIST 2002 SRE show that the spectrum estimation method has a large effect on recognition performance and stabilized weighted LP (SWLP) and minimum variance distortionless response (MVDR) methods yield approximately 7 % and 8 % relative improvements over the standard DFT method at −10 dB SNR level of factory and babble noises, respectively in terms of equal error rate (EER).

Regularization of linear prediction based mel-frequency cepstral coefficient (MFCC) extraction in speaker verification is considered. Commonly, MFCCs are extracted from the discrete Fourier transform (DFT) spectrum of speech frames. In this paper, DFT spectrum estimate is replaced with the recently proposed regularized linear prediction (RLP) method. Regularization of temporally weighted variants, weighted LP (WLP) and stabilized WLP (SWLP) which have earlier shown success in speech and speaker recognition, is also introduced. A novel type of double autocorrelation (DAC) lag windowing is also proposed to enhance robustness. Experiments on the NIST 2002 corpus indicate that regularized all-pole methods (RLP, RWLP and RSWLP) yield large improvement on recognition accuracy under additive factory and babble noise conditions in terms of both equal error rate (EER) and minimum detection cost function (MinDCF).

Speech signals convey various pieces of information such as the identity of its speaker, the language spoken, and the linguistic information about the text being spoken, etc. In this paper, we extract information about the cell phones from their speech records by using mel-frequency cepstrum coefficients and identify their brands and models. Closed-set identification rates of 92.56% and 96.42% have been obtained on a set of 14 different cell phones in the experiments using vector quantization and support vector machine classifiers, respectively.