Speaker Recognition Papers

• Dataset

  • VoxCeleb

    • Voxceleb: A Large-Scale Speaker Identification Dataset

    • Voxceleb2: Deep Speaker Recognition

    • Voxceleb: Large-Scale Speaker Verification In The Wild

  • CnCeleb

    • Cn-Celeb: A Challenging Chinese Speaker Recognition Dataset

    • Cn-Celeb: Multi-Genre Speaker Recognition

• Architecture Design

  • X-Vectors: Robust Dnn Embeddings For Speaker Recognition ( x-vector)

  • But System Description To Voxceleb Speaker Recognition Challenge 2019 (* r-vector*)

  • Rawnet: Advanced End-To-End Deep Neural Network Using Raw Waveforms For Text-Independent Speaker Verification ( RawNet)

  • Speaker Recognition From Raw Waveform With Sincnet (SincNet)

  • Ecapa-Tdnn: Emphasized Channel Attention, Propagation And Aggregation In Tdnn Based Speaker Verification ( ECAPA-TDNN)

• Optimization Objective

  • Classification Based Loss

    • Exploring The Encoding Layer And Loss Function In End-To-End Speaker And Language Recognition System

    • Angular Softmax For Short-Duration Text-Independent Speaker Verification

    • Ensemble Additive Margin Softmax For Speaker Verification

    • Margin Matters: Towards More Discriminative Deep Neural Network Embeddings For Speaker Recognition

    • Large Margin Softmax Loss For Speaker Verification

  • End-to-End Loss

    • End-To-End Text-Dependent Speaker Verification

    • End-To-End Text-Independent Speaker Verification With Triplet Loss On Short Utterances

    • Generalized End-To-End Loss For Speaker Verification

• Pooling Method

  • Attentive Statistics Pooling For Deep Speaker Embedding

  • Multi-Resolution Multi-Head Attention In Deep Speaker Embedding

  • Utterance-Level Aggregation For Speaker Recognition In The Wild

  • A Novel Learnable Dictionary Encoding Layer For End-To-End Language Identification

• Self-supervised Learning

  • Augmentation Adversarial Training For Self-Supervised Speaker Recognition

  • Self-Supervised Text-Independent Speaker Verification Using Prototypical Momentum Contrastive Learning

  • Self-Supervised Speaker Recognition With Loss-Gated Learning