Publication Details
Challenging margin-based speaker embedding extractors by using the variational information bottleneck
Silnova Anna, M.Sc., Ph.D. (DCGM)
Rohdin Johan Andréas, M.Sc., Ph.D. (DCGM)
Plchot Oldřich, Ing., Ph.D. (DCGM)
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
speaker recognition, variational information bottleneck
Speaker embedding extractors are typically trained using a classification loss
over the training speakers. During the last few years, the standard
softmax/cross-entropy loss has been replaced by the margin-based losses, yielding
significant im- provements in speaker recognition accuracy. Motivated by the fact
that the margin merely reduces the logit of the target speaker during training,
we consider a probabilistic framework that has a similar effect. The variational
information bottle- neck provides a principled mechanism for making deterministic
nodes stochastic, resulting in an implicit reduction of the pos- terior of the
target speaker. We experiment with a wide range of speaker recognition benchmarks
and scoring methods and re- port competitive results to those obtained with the
state-of-the- art Additive Angular Margin loss.
@inproceedings{BUT193738,
author="Themos {Stafylakis} and Anna {Silnova} and Johan Andréas {Rohdin} and Oldřich {Plchot} and Lukáš {Burget}",
title="Challenging margin-based speaker embedding extractors by using the variational information bottleneck",
booktitle="Proceedings of Interspeech 2024",
year="2024",
journal="Proceedings of Interspeech",
volume="2024",
number="9",
pages="3220--3224",
publisher="International Speech Communication Association",
address="Kos",
doi="10.21437/Interspeech.2024-2058",
issn="1990-9772",
url="https://www.isca-archive.org/interspeech_2024/stafylakis24_interspeech.pdf"
}