A controlled 120-configuration study of EEND-VC speaker diarization finds finetuned WavLM encoders, Conformer/Mamba decoders, and longer chunks give the biggest gains, with the best system reaching state-of-the-art on five datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Dissecting the Segmentation Model of End-to-End Diarization with Vector Clustering
A controlled 120-configuration study of EEND-VC speaker diarization finds finetuned WavLM encoders, Conformer/Mamba decoders, and longer chunks give the biggest gains, with the best system reaching state-of-the-art on five datasets.