paper-with-me

Papers

Getting More for Less: Using Weak Labels and AV-Mixup for Robust Audio-Visual Speaker Verification

2023-09-13 · Anith Selvakumar, Homa Fashandi

Distance Metric Learning (DML) has typically dominated the audio-visual speaker verification problem space, owing to strong performance in new and unseen classes. In our work, we explored multitask learning techniques to further enhance DML, and show that an auxiliary task with even weak labels can increase the quality of the learned speaker representation without increasing model complexity during inference. We also extend the Generalized End-to-End Loss (GE2E) to multimodal inputs and demonstrate that it can achieve competitive performance in an audio-visual space. Finally, we introduce AV-Mixup, a multimodal augmentation technique during training time that has shown to reduce speaker overfit. Our network achieves state of the art performance for speaker verification, reporting 0.244%, 0.252%, 0.441% Equal Error Rate (EER) on the VoxCeleb1-O/E/H test sets, which is to our knowledge, the best published results on VoxCeleb1-E and VoxCeleb1-H.

📄 PDF Abstract BibTeX arXiv:2309.07115

Code (0)

등록된 구현이 없습니다.

Tasks

Metric LearningMulti-Task LearningSpeaker Verification

Similar Papers 제목 키워드 기반

Adversarial Mixing Policy for Relaxing Locally Linear Constraints in Mixup

2021-09-15 · EMNLP 2021 11 · Guang Liu, Yuzhao Mao, Hailong Huang, Weiguo Gao 외

Mixup is a recent regularizer for current deep classification networks. Through training a neural network on convex combinations of pairs of examples and their labels, it imposes locally linear constraints on the model's…

text-classificationText Classification

CroMo-Mixup: Augmenting Cross-Model Representations for Continual Self-Supervised Learning

2024-07-16 · Erum Mushtaq, Duygu Nur Yaldiz, Yavuz Faruk Bakman, Jie Ding 외

Continual self-supervised learning (CSSL) learns a series of tasks sequentially on the unlabeled data. Two main challenges of continual learning are catastrophic forgetting and task confusion. While CSSL problem has been…

class-incremental learningClass Incremental LearningContinual LearningContinual Self-Supervised Learning+2

DualMix: Unleashing the Potential of Data Augmentation for Online Class-Incremental Learning

2023-03-14 · Yunfeng Fan, Wenchao Xu, Haozhao Wang, Jiaqi Zhu 외

Online Class-Incremental (OCI) learning has sparked new approaches to expand the previously trained model knowledge from sequentially arriving data streams with new classes. Unfortunately, OCI learning can suffer from ca…

class-incremental learningClass Incremental LearningData AugmentationIncremental Learning

Intra-Cluster Mixup: An Effective Data Augmentation Technique for Complementary-Label Learning

2025-09-22 · Tan-Ha Mai, Hsuan-Tien Lin arxiv

In this paper, we investigate the challenges of complementary-label learning (CLL), a specialized form of weakly-supervised learning (WSL) where models are trained with labels indicating classes to which instances do not…

Data Augmentation

Mixup-CAM: Weakly-supervised Semantic Segmentation via Uncertainty Regularization

2020-08-03 · Yu-Ting Chang, Qiaosong Wang, Wei-Chih Hung, Robinson Piramuthu 외

Obtaining object response maps is one important step to achieve weakly-supervised semantic segmentation using image-level labels. However, existing methods rely on the classification task, which could result in a respons…

ClassificationData AugmentationGeneral ClassificationObject+3