mlsanityData to 7 Oct 2026

Bias in Automated Speaker Recognition

arXiv 2201.09486 ↗First posted 24 Jan 2022cs.SD cs.CL cs.CYcited by 20

Not on the map. Why?

Abstract (arXiv)

Automated speaker recognition uses data processing to identify speakers by their voice. Today, automated speaker recognition is deployed on billions of smart devices and in services such as call centres. Despite their wide-scale deployment and known sources of bias in related domains like face recognition and natural language processing, bias in automated speaker recognition has not been studied systematically. We present an in-depth empirical and analytical study of bias in the machine learning development workflow of speaker verification, a voice biometric and core task in automated speaker recognition. Drawing on an established framework for understanding sources of harm in machine learning, we show that bias exists at every development stage in the well-known VoxCeleb Speaker Recognition Challenge, including data generation, model building, and implementation. Most affected are female speakers and non-US nationalities, who experience significant performance degradation. Leveraging the insights from our findings, we make practical recommendations for mitigating bias in automated speaker recognition, and outline future research directions.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 22 of its 61 references that are in the map. First posted on arXiv 24 Jan 2022.

Builds on

22 of 61 references.

References · 22

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  3. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  4. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  5. 2017VoxCeleb: a large-scale speaker identification datasetcited by 714
  6. 2021SUPERB: Speech processing Universal PERformance Benchmarkcited by 652
  7. 2019A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cyclecited by 192
  8. 2020Characterising Bias in Compressed Modelscited by 162
  9. 2020In defence of metric learning for speaker recognitioncited by 123
  10. 2020Speaker Recognition Based on Deep Learning: An Overviewcited by 59
  11. 2017Deep Speaker: an End-to-End Neural Speaker Embedding Systemcited by 56
  12. 2015End-to-End Text-Dependent Speaker Verificationcited by 49
10 more
  1. 2020Clova Baseline System for the VoxCeleb Speaker Recognition Challenge 2020cited by 36
  2. 2020Diversity and Inclusion Metrics in Subset Selectioncited by 27
  3. 2021About Face: A Survey of Facial Recognition Evaluationcited by 27
  4. 2020VoxSRC 2020: The Second VoxCeleb Speaker Recognition Challengecited by 26
  5. 2021VoxCeleb Enrichment for Age and Gender Recognitioncited by 20
  6. 2020The INTERSPEECH 2020 Far-Field Speaker Verification Challengecited by 11
  7. 2019Short-duration Speaker Verification (SdSV) Challenge 2021: the Challenge Evaluation Plancited by 8
  8. 2020Unacceptable, where is my privacy? Exploring Accidental Triggers of Smart Speakerscited by 8
  9. 2022Tiny, always-on and fragile: Bias propagation through design choices in on-device machine learning workflowscited by 8
  10. 2020The Privacy ZEBRA: Zero Evidence Biometric Recognition Assessmentcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 3; 2024: 4; 2025: 3; 2026: 7.

20 papers.

Sort:

  1. 2024Spoken Stereoset: On Evaluating Social Bias Toward Speaker in Speech Large Language Modelscited by 15
  2. 2024FairSSD: Understanding Bias in Synthetic Speech Detectorscited by 9
  3. 2025Forensic deepfake audio detection using segmental speech featurescited by 8
  4. 2026Shortcut Learning in Binary Classifier Black Boxes: Applications to Voice Anti-Spoofing and Biometricscited by 6
  5. 2024Enhancing Supermarket Robot Interaction: A Multi-Level LLM Conversational Interface for Handling Diverse Customer Intentscited by 5
  6. 2022Global Performance Disparities Between English-Language Accents in Automatic Speech Recognitioncited by 4
  7. 2022Design Guidelines for Inclusive Speaker Verification Evaluation Datasetscited by 3
  8. 2022"You Can't Fix What You Can't Measure": Privately Measuring Demographic Performance Disparities in Federated Learningcited by 3
  9. 2023Uncovering Bias in Personal Informaticscited by 3
  10. 2024The Life Cycle of Large Language Models: A Review of Biases in Educationcited by 3
  11. 2023A Study on Bias and Fairness In Deep Speaker Recognitioncited by 2
  12. 2023Using Data Augmentations and VTLN to Reduce Bias in Dutch End-to-End Speech Recognition Systemscited by 2
8 more
  1. 2026Causally Disentangled Contrastive Learning for Multilingual Speaker Embeddingscited by 2
  2. 2026Where Do Self-Supervised Speech Models Become Unfair?cited by 1
  3. 2025A Critical Review of Predominant Bias in Neural Networkscited by 0
  4. 2025Speaker Fuzzy Fingerprints: Benchmarking Text-Based Identification in Multiparty Dialoguescited by 0
  5. 2026LISE : Listenable Interpretable Speaker Embeddingscited by 0
  6. 2026What You Train Is What You Get: Gender Bias, Training Composition, and Post-Hoc Mitigation in Audio Deepfake Detectioncited by 0
  7. 2026When Entanglement Lower-Bounds Disparity: Auditing and Repairing Demographic Fairness in Audio Understanding Modelscited by 0
  8. 2026InterBias-SV: Compound Conditions in Speaker Verificationcited by 0