Token-level logits from a speech foundation model predict ambiguous emotion distributions on IEMOCAP better than its generated text and are more robust to prompt phrasing.
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
-
Token-Level Logits Matter: A Closer Look at Speech Foundation Models for Ambiguous Emotion Recognition
Token-level logits from a speech foundation model predict ambiguous emotion distributions on IEMOCAP better than its generated text and are more robust to prompt phrasing.