A T5-based text-only diarization model reports WDER 4.9 on short dialogues, beating audio baselines, but the result hinges on an uncontrolled in-domain versus out-of-domain comparison.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Do We Still Need Audio? Rethinking Speaker Diarization with a Text-Based Approach Using Multiple Prediction Models
A T5-based text-only diarization model reports WDER 4.9 on short dialogues, beating audio baselines, but the result hinges on an uncontrolled in-domain versus out-of-domain comparison.