A Complete End-to-End Speaker Verification System Using Deep Neural Networks: From Raw Signals to Verification Result

Jung, Jee Weon; Heo, Hee-Soo; Yang, Il-Hyung; Shim, Hye-jin; Yu, Ha-Jin

doi:10.1109/icassp.2018.8462575

Cited by 66 publications

(47 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The b-vector system was first introduced for scoring the i-vectors. Then, it has been shown that the b-vector based system could be efficient classifier of embeddings from DNNs in speaker verification task [6,30]. In our experiments, the b-vector based system comprises three fully-connected lay- ers, each with 512 nodes, and an output layer with two nodes.…”

Section: Additional Validation Using Voxceleb2mentioning

confidence: 99%

End-to-End Losses Based on Speaker Basis Vectors and All-Speaker Hard Negative Mining for Speaker Verification

et al. 2019

Self Cite

View full text Add to dashboard Cite

In recent years, speaker verification has primarily performed using deep neural networks that are trained to output embeddings from input features such as spectrograms or Mel-filterbank energies. Studies that design various loss functions, including metric learning have been widely explored. In this study, we propose two end-to-end loss functions for speaker verification using the concept of speaker bases, which are trainable parameters. One loss function is designed to further increase the interspeaker variation, and the other is designed to conduct the identical concept with hard negative mining. Each speaker basis is designed to represent the corresponding speaker in the process of training deep neural networks. In contrast to the conventional loss functions that can consider only a limited number of speakers included in a mini-batch, the proposed loss functions can consider all the speakers in the training set regardless of the mini-batch composition. In particular, the proposed loss functions enable hard negative mining and calculations of betweenspeaker variations with consideration of all speakers. Through experiments on VoxCeleb1 and VoxCeleb2 datasets, we confirmed that the proposed loss functions could supplement conventional softmax and center loss functions.

show abstract

Section: Additional Validation Using Voxceleb2mentioning

confidence: 99%

End-to-End Losses Based on Speaker Basis Vectors and All-Speaker Hard Negative Mining for Speaker Verification

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…We propose a model (referred to as "RawNet" for convenience) that is an improvement of the CNN-LSTM model in [5,7] by changing architectural details (Section 2.1. ), proposing a modified pre-training scheme (Section 2.2.…”

Section: Front-end: Rawnetmentioning

confidence: 99%

“…It includes a number of modifications to the CNN-LSTM model in [5,7], allowing for further improvement. First, activation functions are changed from rectified linear units (ReLU) to leaky ReLU.…”

Section: Model Architecturementioning

confidence: 99%

“…We didn't apply any pre-processing, such as normalization, except pre-emphasis [30] to raw waveforms which were experimentally shown effective in our internal comparison experiments. For mini-batch construction, utterances were either cropped or duplicated (concatenated until the length reach) into 59049 samples (≈ 3.59s) in the training phase, following [5,7]. In the evaluation phase, no adjustments were made to length; the whole utterance was used.…”

Section: Experimental Configurationsmentioning

confidence: 99%

See 1 more Smart Citation

RawNet: Advanced End-to-End Deep Neural Network Using Raw Waveforms for Text-Independent Speaker Verification

et al. 2019

Self Cite

View full text Add to dashboard Cite

Recently, direct modeling of raw waveforms using deep neural networks has been widely studied for a number of tasks in audio domains. In speaker verification, however, utilization of raw waveforms is in its preliminary phase, requiring further investigation. In this study, we explore end-to-end deep neural networks that input raw waveforms to improve various aspects: front-end speaker embedding extraction including model architecture, pre-training scheme, additional objective functions, and back-end classification. Adjustment of model architecture using a pre-training scheme can extract speaker embeddings, giving a significant improvement in performance. Additional objective functions simplify the process of extracting speaker embeddings by merging conventional two-phase processes: extracting utterance-level features such as i-vectors or x-vectors and the feature enhancement phase, e.g., linear discriminant analysis. Effective back-end classification models that suit the proposed speaker embedding are also explored. We propose an end-toend system that comprises two deep neural networks, one frontend for utterance-level speaker embedding extraction and the other for back-end classification. Experiments conducted on the VoxCeleb1 dataset demonstrate that the proposed model achieves state-of-the-art performance among systems without data augmentation. The proposed system is also comparable to the state-of-the-art x-vector system that adopts data augmentation.

show abstract

“…The DNN used in this study comprises convolutional neural networks (CNNs), gated recurrent units (GRUs) and fully connected layers (CNN-GRU) as used in [15][16][17]. In this architecture, input features are first processed using convolutional layers to extract frame-level embeddings.…”

Section: End-to-end Dnnmentioning

confidence: 99%

Replay Attack Detection with Complementary High-Resolution Information Using End-to-End DNN for the ASVspoof 2019 Challenge

Jung¹,

Shim²,

Heo³

et al. 2019

Interspeech 2019

Self Cite

View full text Add to dashboard Cite

In this study, we concentrate on replacing the process of extracting hand-crafted acoustic feature with end-to-end DNN using complementary high-resolution spectrograms. As a result of advance in audio devices, typical characteristics of a replayed speech based on conventional knowledge alter or diminish in unknown replay configurations. Thus, it has become increasingly difficult to detect spoofed speech with a conventional knowledge-based approach. To detect unrevealed characteristics that reside in a replayed speech, we directly input spectrograms into an end-to-end DNN without knowledge-based intervention. Explorations dealt in this study that differentiates from existing spectrogram-based systems are twofold: complementary information and high-resolution. Spectrograms with different information are explored, and it is shown that additional information such as the phase information can be complementary. High-resolution spectrograms are employed with the assumption that the difference between a bona-fide and a replayed speech exists in the details. Additionally, to verify whether other features are complementary to spectrograms, we also examine raw waveform and an i-vector based system. Experiments conducted on the ASVspoof 2019 physical access challenge show promising results, where t-DCF and equal error rates are 0.0570 and 2.45 % for the evaluation set, respectively.

show abstract

A Complete End-to-End Speaker Verification System Using Deep Neural Networks: From Raw Signals to Verification Result

Cited by 66 publications

References 10 publications

End-to-End Losses Based on Speaker Basis Vectors and All-Speaker Hard Negative Mining for Speaker Verification

End-to-End Losses Based on Speaker Basis Vectors and All-Speaker Hard Negative Mining for Speaker Verification

RawNet: Advanced End-to-End Deep Neural Network Using Raw Waveforms for Text-Independent Speaker Verification

Replay Attack Detection with Complementary High-Resolution Information Using End-to-End DNN for the ASVspoof 2019 Challenge

Contact Info

Product

Resources

About