Cosine Scoring With Uncertainty for Neural Speaker Embedding

Page view(s)
Checked on Jul 10, 2024
Cosine Scoring With Uncertainty for Neural Speaker Embedding
Cosine Scoring With Uncertainty for Neural Speaker Embedding
Journal Title:
IEEE Signal Processing Letters
Publication Date:
08 March 2024
Wang, Q., & Lee, K. A. (2024). Cosine Scoring With Uncertainty for Neural Speaker Embedding. IEEE Signal Processing Letters, 31, 845–849.
Uncertainty modeling in speaker representation aims to learn the variability present in speech utterances. While the conventional cosine-scoring is computationally efficient and prevalent in speaker recognition, it lacks the capability to handle uncertainty. To address this challenge, this paper proposes an approach for estimating uncertainty at the speaker embedding front-end and propagating it to the cosine scoring back-end. Experiments conducted on the VoxCeleb and SITW datasets confirmed the efficacy of the proposed method in handling uncertainty arising from embedding estimation. It achieved improvement with 8.5% and 9.8% average reductions in EER and minDCF compared to the conventional cosine similarity. It is also computationally efficient in practice.
License type:
Publisher Copyright
Funding Info:
There was no specific funding for the research done
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Files uploaded:

File Size Format Action
paper-new.pdf 495.31 KB PDF Request a copy