# Xi-Vector Embedding for Speaker Recognition

arXiv: https://arxiv.org/abs/2108.05679 · HTML: https://mlsanity.com/paper/2108.05679/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Aug 2021.

## Built on by (most cited first, 9 of 9)

- [Disentangling Voice and Content with Self-Supervision for Speaker Recognition](https://mlsanity.com/paper/2310.01128/) (2023)
- [MFA: TDNN with Multi-scale Frequency-channel Attention for Text-independent Speaker Verification with Short Utterances](https://mlsanity.com/paper/2202.01624/) (2022)
- [Audio-Visual Speaker Verification via Joint Cross-Attention](https://mlsanity.com/paper/2309.16569/) (2023)
- [Asynchronous Voice Anonymization Using Adversarial Perturbation On Speaker Embedding](https://mlsanity.com/paper/2406.08200/) (2024)
- [Interpolating Speaker Identities in Embedding Space for Data Expansion](https://mlsanity.com/paper/2508.19210/) (2025)
- [I4U System Description for NIST SRE'20 CTS Challenge](https://mlsanity.com/paper/2211.01091/) (2022)
- [Cosine Scoring with Uncertainty for Neural Speaker Embedding](https://mlsanity.com/paper/2403.06404/) (2024)
- [Kiwano: A Cutting-Edge Open-Source Toolkit for Speaker Verification](https://mlsanity.com/paper/2606.22369/) (2026)
- [DEFT: Disentanglement-Enhanced Fine-Tuning for EEG Foundation Models](https://mlsanity.com/paper/2610.14497/) (2026)

## Abstract (arXiv)

We present a Bayesian formulation for deep speaker embedding, wherein the xi-vector is the Bayesian counterpart of the x-vector, taking into account the uncertainty estimate. On the technology front, we offer a simple and straightforward extension to the now widely used x-vector. It consists of an auxiliary neural net predicting the frame-wise uncertainty of the input sequence. We show that the proposed extension leads to substantial improvement across all operating points, with a significant reduction in error rates and detection cost. On the theoretical front, our proposal integrates the Bayesian formulation of linear Gaussian model to speaker-embedding neural networks via the pooling layer. In one sense, our proposal integrates the Bayesian formulation of the i-vector to that of the x-vector. Hence, we refer to the embedding as the xi-vector, which is pronounced as /zai/ vector. Experimental results on the SITW evaluation set show a consistent improvement of over 17.5% in equal-error-rate and 10.9% in minimum detection cost.