MATER: Multi-level Acoustic and Textual Emotion Representation for Interpretable Speech Emotion Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jon, Hyo Jin, Jin, Longbin, Jung, Hyuntaek, Kim, Hyunseo, Min, Donghun, Kim, Eun Yi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908590563917824
author Jon, Hyo Jin
Jin, Longbin
Jung, Hyuntaek
Kim, Hyunseo
Min, Donghun
Kim, Eun Yi
author_facet Jon, Hyo Jin
Jin, Longbin
Jung, Hyuntaek
Kim, Hyunseo
Min, Donghun
Kim, Eun Yi
contents This paper presents our contributions to the Speech Emotion Recognition in Naturalistic Conditions (SERNC) Challenge, where we address categorical emotion recognition and emotional attribute prediction. To handle the complexities of natural speech, including intra- and inter-subject variability, we propose Multi-level Acoustic-Textual Emotion Representation (MATER), a novel hierarchical framework that integrates acoustic and textual features at the word, utterance, and embedding levels. By fusing low-level lexical and acoustic cues with high-level contextualized representations, MATER effectively captures both fine-grained prosodic variations and semantic nuances. Additionally, we introduce an uncertainty-aware ensemble strategy to mitigate annotator inconsistencies, improving robustness in ambiguous emotional expressions. MATER ranks fourth in both tasks with a Macro-F1 of 41.01% and an average CCC of 0.5928, securing second place in valence prediction with an impressive CCC of 0.6941.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19887
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MATER: Multi-level Acoustic and Textual Emotion Representation for Interpretable Speech Emotion Recognition
Jon, Hyo Jin
Jin, Longbin
Jung, Hyuntaek
Kim, Hyunseo
Min, Donghun
Kim, Eun Yi
Audio and Speech Processing
Artificial Intelligence
Sound
68T10
This paper presents our contributions to the Speech Emotion Recognition in Naturalistic Conditions (SERNC) Challenge, where we address categorical emotion recognition and emotional attribute prediction. To handle the complexities of natural speech, including intra- and inter-subject variability, we propose Multi-level Acoustic-Textual Emotion Representation (MATER), a novel hierarchical framework that integrates acoustic and textual features at the word, utterance, and embedding levels. By fusing low-level lexical and acoustic cues with high-level contextualized representations, MATER effectively captures both fine-grained prosodic variations and semantic nuances. Additionally, we introduce an uncertainty-aware ensemble strategy to mitigate annotator inconsistencies, improving robustness in ambiguous emotional expressions. MATER ranks fourth in both tasks with a Macro-F1 of 41.01% and an average CCC of 0.5928, securing second place in valence prediction with an impressive CCC of 0.6941.
title MATER: Multi-level Acoustic and Textual Emotion Representation for Interpretable Speech Emotion Recognition
topic Audio and Speech Processing
Artificial Intelligence
Sound
68T10
url https://arxiv.org/abs/2506.19887