Abhinaya, an ensemble of fine-tuned SSL, SLLM, and LLM models with majority voting, achieved state-of-the-art macro-F1 (44.02%) on the Interspeech 2025 naturalistic speech emotion recognition test set.
neutral”) almost26 times that of the minority class (“fear
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
-
ABHINAYA -- A System for Speech Emotion Recognition In Naturalistic Conditions Challenge
Abhinaya, an ensemble of fine-tuned SSL, SLLM, and LLM models with majority voting, achieved state-of-the-art macro-F1 (44.02%) on the Interspeech 2025 naturalistic speech emotion recognition test set.