Covariance Regularization for Probabilistic Linear Discriminant Analysis
Preprint
Abstract
Probabilistic linear discriminant analysis (PLDA) is commonly used in speaker verification systems to score the similarity of speaker embeddings. Recent studies improved the performance of PLDA in domain-matched conditions by diagonalizing its covariance. We suspect such brutal pruning approach could eliminate its capacity in modeling dimension correlation of speaker embeddings, leading to inadequate performance with domain adaptation. This paper explores two alternative covariance regularization approaches, namely, interpolated PLDA and sparse PLDA, to tackle the problem. The interpolated PLDA incorporates the prior knowledge from cosine scoring to interpolate the covariance of PLDA. The sparse PLDA introduces a sparsity penalty to update the covariance. Experimental results demonstrate that both approaches outperform diagonal regularization noticeably with domain adaptation. In addition, in-domain data can be significantly reduced when training sparse PLDA for domain adaptation.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- CN-Celeb: Multi-genre speaker recognitionSpeech Communication, 2022
- Unsupervised Bayesian Adaptation of PLDA for Speaker Verification2021
- ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification2020
- On Robustness of Unsupervised Domain Adaptation for Speaker Recognition2019
- Large Margin Softmax Loss for Speaker Verification2019
- VoxCeleb2: Deep Speaker Recognition2018
- Additive Margin Softmax for Face VerificationIEEE Signal Processing Letters, 2018
- Deep Neural Network Embeddings for Text-Independent Speaker Verification2017
- VoxCeleb: A Large-Scale Speaker Identification Dataset2017
- Analysis of i-vector length normalization in speaker recognition systems2011