Triplet-Center Loss Based Deep Embedding Learning Method for Speaker Verification

Jiang, Yiheng; Yan, Shuicheng; Yan, Jie; Dai, Li-Rong; McLoughlin, Ian

doi:10.1109/apsipaasc47483.2019.9023301

Cited by 3 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Center loss 39 only used to reduce the intra‐class differences, cannot effectively increase the inter‐class differences. In recent years, the idea of combing triplet loss and center loss (named triplet center loss) to solve 3D visual analysis 40 and speaker verification 41 shows high accuracy compared with other loss. However, our method has lower time complexity compared with the existing methods.…”

Section: Proposed Methodsmentioning

confidence: 99%

Centralized contrastive loss with weakly supervised progressive feature extraction for fine‐grained common thorax disease retrieval in chest x‐ray

et al. 2023

View full text Add to dashboard Cite

Background Medical images have already become an essential tool for the diagnosis of many diseases. Thus a large number of medical images are being generated due to the daily routine inspection. An efficient image‐based disease retrieval system will not only make full use of existing data, but also help physicians to prognosis the diseases. Medical image retrieval is represented by the classification and localization of common thorax diseases in x‐ray images. Although extensive efforts have been put into this field, there are still many challenges. Purpose Most of the existing fine‐grained image research methods just apply existing deep learning frameworks in extracting the image features. However, these high‐level features mainly focus on the global representations of the object, rather than simultaneously considering the local ones. It requires fine‐grained details to classify the images with similar lesion areas. Thus, it is necessary to combine the global features and local ones to make the features more discriminative. On the other hand, training CNN models based on current existing strategies have a high time complexity, and is hard to get the discriminative features mentioned above. In addition, the visual retrieval method of fine‐grained medical images still has the problem of insufficient sample data with accurate annotation information. Methods To address above challenges, we introduced a novel fine‐grained medical images retrieval method. First, a centralized contrastive loss (CCLoss) is proposed as our metric learning loss function. Parameters are updated by using the center point, which not only improves the distinguishing performance of features, but also effectively reduces the time complexity of the algorithm. In addition, a weakly supervised progressive feature extraction method is proposed to gradually extract the combined features. And the attention mechanism module is applied to screen the target information after the initial positioning for fine refinement, so as to separate the features with a high degree of discrimination. The retrieval of 14 different chest diseases is evaluated on the chest x‐ray datasets. Results Compared with the existing research methods, the proposed method shows a better retrieval result for Recall@8 by 2.26%∼4.6%$\%{\sim }4.6\%$ and achieves a very efficient training speed which is 100 times faster than the pair‐wise loss‐based training strategy. We also assessed the effects of Recall@k (k = 2, 4, 6, 8) for progressive features extracted from different steps to obtain a model with the best retrieval performance. Conclusions The proposed model is capable of learning discriminative representations from chest x‐ray datasets, and it achieves better performance compared with other state‐of‐the‐art methods. Therefore, the developed model would be useful in the diagnosis of common thorax disease or unknown chest disease.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Centralized contrastive loss with weakly supervised progressive feature extraction for fine‐grained common thorax disease retrieval in chest x‐ray

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Different network structures have been used for framelevel extraction in recent studies, including the time-delay neural network (TDNN) [2], convolutional neural network (CNN) [5,8] and Long Short-Term Memory (LSTM) network [9]. Meanwhile many authors [5,10,11,12] have investigated residual networks (ResNet) [13] with different settings (i.e. width and depth) as a backbone for embedding learning, which can effectively incorporate the information from previous layers with skip connection.…”

Section: Introductionmentioning

confidence: 99%

An Effective Deep Embedding Learning Method Based on Dense-Residual Networks for Speaker Verification

Liu

Yan

McLoughlin

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

In this paper, we present an effective end-to-end deep embedding learning method based on Dense-Residual networks, which combine the advantages of a densely connected convolutional network (DenseNet) and a residual network (ResNet), for speaker verification (SV). Unlike a model ensemble strategy which merges the results of multiple systems, the proposed Dense-Residual networks perform feature fusion on every basic DenseR building block. Specifically, two types of DenseR blocks are designed. A sequential-DenseR block is constructed by densely connecting stacked basic units in a residual block of ResNet. A parallel-DenseR comprises split and concatenation operations on residual and dense components via corresponding skip connections. These building blocks are stacked into deep networks to exploit the complementary information with different receptive field sizes and growth rates. Extensive experiments have been conducted on the VoxCeleb1 dataset to evaluate the proposed methods. The SV performance achieved by the proposed Dense-Residual networks is shown to outperform corresponding ResNet, DenseNet or fusions of them, with similar model complexity, by a significant margin.

show abstract

“…Many recent works have focused on utterance-level aggregation methods, e.g., statistical pooling [3], attentive pooling [8,9], bilinear pooling [4], and dictionary based pooling methods [10,11]. Meanwhile, other works have proposed different loss functions, including triplet loss [12,13], center loss [10], triplet-center loss [14], angular softmax loss (A-Softmax) [10] and additive margin softmax loss (AM-Softmax) [15,16]. However, in most deep embedding learning methods, the network architectures are trained under identification supervision, optimized for the SID task.…”

Section: Introductionmentioning

confidence: 99%

An Effective Speaker Recognition Method Based on Joint Identification and Verification Supervisions

et al. 2020

View full text Add to dashboard Cite

Deep embedding learning based speaker verification methods have attracted significant recent research interest due to their superior performance. Existing methods mainly focus on designing frame-level feature extraction structures, utterance-level aggregation methods and loss functions to learn discriminative speaker embeddings. The scores of verification trials are then computed using cosine distance or Probabilistic Linear Discriminative Analysis (PLDA) classifiers. This paper proposes an effective speaker recognition method which is based on joint identification and verification supervisions, inspired by multi-task learning frameworks. Specifically, a deep architecture with convolutional feature extractor, attentive pooling and two classifier branches is presented. The first, an identification branch, is trained with additive margin softmax loss (AM-Softmax) to classify the speaker identities. The second, a verification branch, trains a discriminator with binary cross entropy loss (BCE) to optimize a new triplet-based mutual information. To balance the two losses during different training stages, a ramp-up/ramp-down weighting scheme is employed. Furthermore, an attentive bilinear pooling method is proposed to improve the effectiveness of embeddings. Extensive experiments have been conducted on VoxCeleb1 to evaluate the proposed method, demonstrating results that relatively reduce the equal error rate (EER) by 22% compared to the baseline system using identification supervision only.

show abstract

Triplet-Center Loss Based Deep Embedding Learning Method for Speaker Verification

Cited by 3 publications

References 15 publications

Centralized contrastive loss with weakly supervised progressive feature extraction for fine‐grained common thorax disease retrieval in chest x‐ray

Centralized contrastive loss with weakly supervised progressive feature extraction for fine‐grained common thorax disease retrieval in chest x‐ray

An Effective Deep Embedding Learning Method Based on Dense-Residual Networks for Speaker Verification

An Effective Speaker Recognition Method Based on Joint Identification and Verification Supervisions

Contact Info

Product

Resources

About