Blueprint Separable Subsampling and Aggregate Feature Conformer-Based End-to-End Neural Diarization

Blueprint Separable Subsampling and Aggregate Feature Conformer-Based End-to-End Neural Diarization

At present, a prevalent approach to speaker diarization is clustering based on speaker embeddings. However, this method encounters two primary issues. Firstly, it cannot directly minimize the diarization error during the training process; secondly, the majority of clustering-based methods struggle t...

Full description

Bibliographic Details
Main Authors: Xiaolin Jiao, Yaqi Chen, Dan Qu, Xukui Yang
Format: Article
Language:English
Published: MDPI AG 2023-10-01
Series:Electronics
Subjects:
end-to-end neural diarization (EEND)
blueprint separable convolution (BSConv)
multi-scale feature aggregation (MFA)
Online Access:https://www.mdpi.com/2079-9292/12/19/4118

Similar Items

Similar Items