Ensemble Models for Spoofing Detection in Automatic Speaker Verification

Chettri, Bhusan; Stoller, Daniel; Morfi, Veronica; Ramírez, M. A.; Benetos, Emmanouil; Sturm, Bob L.

doi:10.48550/arxiv.1904.04589

Cited by 7 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Confirming the observations made in [9] and [23], the input features are not normalized, but simply scaled down to be in the range from −1 to 1.…”

Section: Audio Preprocessing and Feature Extractionmentioning

confidence: 80%

“…Based on existing literature (e.g. [23]), it can be explained that the beginning and tailing silence cues can lead to better performance. Considering these findings and our practical application, we decided to use 8.5s input length and to do cutting and padding at the end of the audio from now on (so that we do not rely on voice activity detection in practical applications).…”

Section: Experimental Setup and Resultsmentioning

confidence: 99%

“…LFBANK subsequently applies 80 filters (cf. with [23]) without any delta coefficients. The resulting input dimension for GD gram/LOGSPEC and LFBANK is 401 × 566 and 80 × 566, respectively.…”

Section: Audio Preprocessing and Feature Extractionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Task Siamese Neural Network for Improving Replay Attack Detection

Platen¹,

Tao²,

Tür³

2020

Preprint

View full text Add to dashboard Cite

Automatic speaker verification systems are vulnerable to audio replay attacks which bypass security by replaying recordings of authorized speakers. Replay attack detection (RA) detection systems built upon Residual Neural Networks (ResNet)s have yielded astonishing results on the public benchmark ASVspoof 2019 Physical Access challenge. With most teams using fine-tuned feature extraction pipelines and model architectures, the generalizability of such systems remains questionable though. In this work, we analyse the effect of discriminative feature learning in a multi-task learning (MTL) setting can have on the generalizability and discriminability of RA detection systems. We use a popular ResNet architecture optimized by the cross-entropy criterion as our baseline and compare it to the same architecture optimized by MTL using Siamese Neural Networks (SNN). It can be shown that SNN outperform the baseline by relative 26.8% Equal Error Rate (EER). We further enhance the model's architecture and demonstrate that SNN with additional reconstruction loss yield another significant improvement of relative 13.8 % EER.

show abstract

“…Confirming the observations made in [9] and [23], the input features are not normalized, but simply scaled down to be in the range from −1 to 1.…”

Section: Audio Preprocessing and Feature Extractionmentioning

confidence: 80%

Section: Experimental Setup and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Task Siamese Neural Network for Improving Replay Attack Detection

Platen¹,

Tao²,

Tür³

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…There is some previous work that examines the impact of silence in ASVspoof. For example, [8] find that the length of silence in the PA part of ASVspoof 2019 differs between bonafide and spoof data. Additionally, for the ASVspoof 2017 replay detection challenge, [14] observe that the presence of certain attributes such as 'clicks', trailing and leading silence in the audio may hint at the target label.…”

Section: Related Workmentioning

confidence: 97%

Speech is Silver, Silence is Golden: What do ASVspoof-trained Models Really Learn?

Müller¹,

Dieckmann²,

Pavel³

et al. 2021

2021 Edition of the Automatic Speaker Verification and Spoofing Countermeasures Challenge

View full text Add to dashboard Cite

The ASVspoof Dataset is one of the most established datasets for training and benchmarking systems designed for the detection of spoofed audio and audio deepfakes. However, we observe an uneven distribution of silence length in dataset's training and test data, which hints at the target label: Bona-fide instances tend to have significantly longer leading + trailing silences than spoofed instances. This could be problematic, since a model may learn to only, or at least partially, base its decision on the length of the silence (similar to the issue with the Pascal VOC 2007 dataset, where all images of horses also contained a specific watermark [1]). In this paper, we explore this phenomenon in depth. We train a number of networks on only a) the length of the leading silence and b) with and without leading + trailing silence. Results show that models trained on only the length of the leading silence perform suspiciously well: They achieve up to 85% percent accuracy and an equal error rate (EER) of 0.15 on the 'eval' split of the data. Conversely, when training strong models on the full audio files, we observe that trimming silence during preprocessing dramatically worsens performance (EER increases from 0.03 to 0.15). This could indicate that previous work may, in part, have learned only to classify targets based on the length of silence. Consequently, it could mean that spoofing detection may not be as advanced as previous high-scores have led to believe. We hope that by sharing these results, the ASV community can further evaluate this phenomenon.

show abstract

“…The recurrent ASVspoof contest challenges researchers to find suitable machine learning algorithms for this task. There is a large body of related work [10,11,12,13], which uses machine learning to identify artifacts in audio waveforms that may indicate a deepfake. Such artifacts include noisy glitch, phase mismatch, reverberation, or loss of intelligibility [14,15], but also artefacts that humans cannot perceive.…”

Section: Introductionmentioning

confidence: 99%

Human Perception of Audio Deepfakes

Müller¹,

Markert²,

Williams³

2021

Preprint

View full text Add to dashboard Cite

The recent emergence of deepfakes, computerized realistic multimedia fakes, brought the detection of manipulated and generated content to the forefront. While many machine learning models for deepfakes detection have been proposed, the human detection capabilities have remained far less explored. This is of special importance as human perception differs from machine perception and deepfakes are generally designed to fool the human. So far, this issue has only been addressed in the area of images and video.To compare the ability of humans and machines in detecting audio deepfakes, we conducted an online gamified experiment in which we asked users to discern bonda-fide audio samples from spoofed audio, generated with a variety of algorithms. 200 users competed for 8967 game rounds with an artificial intelligence (AI) algorithm trained for audio deepfake detection. With the collected data we found that the machine generally outperforms the humans in detecting audio deepfakes, but that the converse holds for a certain attack type, for which humans are still more accurate. Furthermore, we found that younger participants are on average better at detecting audio deepfakes than older participants, while IT-professionals hold no advantage over laymen. We conclude that it is important to combine human and machine knowledge in order to improve audio deepfake detection.

show abstract

Ensemble Models for Spoofing Detection in Automatic Speaker Verification

Cited by 7 publications

References 0 publications

Multi-Task Siamese Neural Network for Improving Replay Attack Detection

Multi-Task Siamese Neural Network for Improving Replay Attack Detection

Speech is Silver, Silence is Golden: What do ASVspoof-trained Models Really Learn?

Human Perception of Audio Deepfakes

Contact Info

Product

Resources

About