Meta-PerSER uses MAML-style meta-training with combined-set training, derivative annealing, and per-layer learning rates to personalize speech emotion recognition to unseen annotators from 32 labeled examples, outperforming fine-tuning and multi-task baselines on IEMOCAP.
Backbone SER Framework We employ a unified model architecture based on the s3prl toolkit [25]
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Meta-PerSER: Few-Shot Listener Personalized Speech Emotion Recognition via Meta-learning
Meta-PerSER uses MAML-style meta-training with combined-set training, derivative annealing, and per-layer learning rates to personalize speech emotion recognition to unseen annotators from 32 labeled examples, outperforming fine-tuning and multi-task baselines on IEMOCAP.