# A Study on Bias and Fairness In Deep Speaker Recognition

arXiv: https://arxiv.org/abs/2303.08026 · HTML: https://mlsanity.com/paper/2303.08026/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 14 Mar 2023.

## Builds on (most cited first, 20 of 20)

- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [In defence of metric learning for speaker recognition](https://mlsanity.com/paper/2003.11982/) (2020)
- [Return of the Devil in the Details: Delving Deep into Convolutional Nets](https://mlsanity.com/paper/1405.3531/) (2014)
- [Speaker Recognition Based on Deep Learning: An Overview](https://mlsanity.com/paper/2012.00931/) (2020)
- [Pushing the limits of raw waveform speaker recognition](https://mlsanity.com/paper/2203.08488/) (2022)
- [Exploring the Encoding Layer and Loss Function in End-to-End Speaker and Language Recognition System](https://mlsanity.com/paper/1804.05160/) (2018)
- [Personalization Strategies for End-to-End Speech Recognition Systems](https://mlsanity.com/paper/2102.07739/) (2021)
- [The ins and outs of speaker recognition: lessons from VoxSRC 2020](https://mlsanity.com/paper/2010.15809/) (2020)
- [Bias in Automated Speaker Recognition](https://mlsanity.com/paper/2201.09486/) (2022)
- [Delving into VoxCeleb: environment invariant speaker recognition](https://mlsanity.com/paper/1910.11238/) (2019)
- [Improving fairness in speaker verification via Group-adapted Fusion Network](https://mlsanity.com/paper/2202.11323/) (2022)
- [Siamese Capsule Network for End-to-End Speaker Recognition In The Wild](https://mlsanity.com/paper/2009.13480/) (2020)
- [A Deep Neural Network for Short-Segment Speaker Recognition](https://mlsanity.com/paper/1907.10420/) (2019)
- [Improving Fairness in Speaker Recognition](https://mlsanity.com/paper/2104.14067/) (2021)
- [SVEva Fair: A Framework for Evaluating Fairness in Speaker Verification](https://mlsanity.com/paper/2107.12049/) (2021)
- [Fine-grained Early Frequency Attention for Deep Speaker Recognition](https://mlsanity.com/paper/2207.10006/) (2022)

## Built on by (most cited first, 2 of 2)

- [SVeritas: Benchmark for Robust Speaker Verification under Diverse Conditions](https://mlsanity.com/paper/2509.17091/) (2025)
- [Shortcut Learning in Binary Classifier Black Boxes: Applications to Voice Anti-Spoofing and Biometrics](https://mlsanity.com/paper/2601.17782/) (2026)

## Abstract (arXiv)

With the ubiquity of smart devices that use speaker recognition (SR) systems as a means of authenticating individuals and personalizing their services, fairness of SR systems has becomes an important point of focus. In this paper we study the notion of fairness in recent SR systems based on 3 popular and relevant definitions, namely Statistical Parity, Equalized Odds, and Equal Opportunity. We examine 5 popular neural architectures and 5 commonly used loss functions in training SR systems, while evaluating their fairness against gender and nationality groups. Our detailed experiments shed light on this concept and demonstrate that more sophisticated encoder architectures better align with the definitions of fairness. Additionally, we find that the choice of loss functions can significantly impact the bias of SR models.