mlsanityData to 7 Oct 2026

In defence of metric learning for speaker recognition

arXiv 2003.11982 ↗First posted 26 Mar 2020eess.AS cs.SDcited by 123

Not on the map. Why?

Abstract (arXiv)

The objective of this paper is 'open-set' speaker recognition of unseen speakers, where ideal embeddings should be able to condense information into a compact utterance-level representation that has small intra-speaker and large inter-speaker distance. A popular belief in speaker recognition is that networks trained with classification objectives outperform metric learning methods. In this paper, we present an extensive evaluation of most popular loss functions for speaker recognition on the VoxCeleb dataset. We demonstrate that the vanilla triplet loss shows competitive performance compared to classification-based losses, and those trained with our proposed metric learning objective outperform state-of-the-art methods.

Key facts. Cited by 123 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 25; 2022: 29; 2023: 33; 2024: 19; 2025: 7; 2026: 10.

123 papers.

Sort:

By area:

  1. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,530
  2. 2021Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representationcited by 288
  3. 2021YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyonecited by 277
  4. 2022Learning Hierarchical Cross-Modal Association for Co-Speech Gesture Generationcited by 106
  5. 2022Talking Head Generation with Probabilistic Audio-to-Visual Diffusion Priorscited by 55
  6. 2023EmotionGesture: Audio-Driven Diverse Emotional Co-Speech 3D Gesture Generationcited by 40
  7. 2023MuAViC: A Multilingual Audio-Visual Corpus for Robust Speech Recognition and Robust Speech-to-Text Translationcited by 39
  8. 2023HierSpeech++: Bridging the Gap between Semantic and Acoustic Representation of Speech by Hierarchical Variational Inference for Zero-shot Speech Synthesiscited by 35
  9. 2021SynthASR: Unlocking Synthetic Data for Speech Recognitioncited by 34
  10. 2021Face, Body, Voice: Video Person-Clustering with Multiple Modalitiescited by 24
  11. 2021AVA-AVD: Audio-Visual Speaker Diarization in the Wildcited by 24
  12. 2024ESPnet-SPK: full pipeline speaker embedding toolkit with reproducible recipes, self-supervised front-ends, and off-the-shelf modelscited by 22
38 more
  1. 2022Bias in Automated Speaker Recognitioncited by 20
  2. 2022Detecting Individual Decision-Making Style: Exploring Behavioral Stylometry in Chesscited by 18
  3. 2021Attack on practical speaker verification system using universal adversarial perturbationscited by 17
  4. 2023Weakly-Supervised Emotion Transition Learning for Diverse 3D Co-speech Gesture Generationcited by 17
  5. 2024Audio-Visual Speech Representation Expert for Enhanced Talking Face Video Generation and Evaluationcited by 17
  6. 2022Domain Generalization with Relaxed Instance Frequency-wise Normalization for Multi-device Acoustic Scene Classificationcited by 15
  7. 2021Self-Supervised Speaker Verification with Simple Siamese Network and Self-Supervised Regularizationcited by 14
  8. 2023Lexical Speaker Error Correction: Leveraging Language Models for Speaker Diarization Error Correctioncited by 14
  9. 2023SLMIA-SR: Speaker-Level Membership Inference Attacks against Speaker Recognition Systemscited by 14
  10. 2023Audio-driven Talking Face Generation with Stabilized Synchronization Losscited by 13
  11. 2021Voting for the right answer: Adversarial defense for speaker verificationcited by 12
  12. 2022MFA: TDNN with Multi-scale Frequency-channel Attention for Text-independent Speaker Verification with Short Utterancescited by 12
  13. 2022Deepfake audio detection by speaker verificationcited by 11
  14. 2023AV-MaskEnhancer: Enhancing Video Representations through Audio-Visual Masked Autoencodercited by 11
  15. 2022Frequency and Multi-Scale Selective Kernel Attention for Speaker Verificationcited by 10
  16. 2022Label-Efficient Self-Supervised Speaker Verification With Information Maximization and Contrastive Learningcited by 10
  17. 2023Zero-shot text-to-speech synthesis conditioned using self-supervised speech representation modelcited by 10
  18. 2023Disentangling Voice and Content with Self-Supervision for Speaker Recognitioncited by 10
  19. 2024Exploring WavLM Back-ends for Speech Spoofing and Deepfake Detectioncited by 10
  20. 2021Automated Video Labelling: Identifying Faces by Corroborative Evidencecited by 9
  21. 2023Few-Shot Open-Set Learning for On-Device Customization of KeyWord Spotting Systemscited by 9
  22. 2024The VoxCeleb Speaker Recognition Challenge: A Retrospectivecited by 9
  23. 2021Adversarial Sample Detection for Speaker Verification by Neural Vocoderscited by 8
  24. 2022Learning Lip-Based Audio-Visual Speaker Embeddings with AV-HuBERTcited by 8
  25. 2023VoxSRC 2022: The Fourth VoxCeleb Speaker Recognition Challengecited by 8
  26. 2022Using Active Speaker Faces for Diarization in TV showscited by 7
  27. 2022Personalized Keyword Spotting through Multi-task Learningcited by 7
  28. 2023Towards single integrated spoofing-aware speaker verification embeddingscited by 7
  29. 2024An Effective Automated Speaking Assessment Approach to Mitigating Data Scarcity and Imbalanced Distributioncited by 7
  30. 2022Attacker Attribution of Audio Deepfakescited by 6
  31. 2024Singer Identity Representation Learning using Self-Supervised Techniquescited by 6
  32. 2021Protecting gender and identity with disentangled speech representationscited by 5
  33. 2021Directed Speech Separation for Automatic Speech Recognition of Long Form Conversational Speechcited by 5
  34. 2022Generating gender-ambiguous voices for privacy-preserving speech recognitioncited by 5
  35. 2022Multi-Frequency Information Enhanced Channel Attention Module for Speaker Representation Learningcited by 5
  36. 2021Adapting Speaker Embeddings for Speaker Diarisationcited by 4
  37. 2022Contrastive-mixup learning for improved speaker verificationcited by 4
  38. 2022Domain Agnostic Few-shot Learning for Speaker Verificationcited by 4

50 most cited of 123 All 123 (JSON)