MSF-SER: Enriching Acoustic Modeling with Multi-Granularity Semantics for Speech Emotion Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Haoxun, Sun, Yuqing, Shi, Hanlei, Liu, Yu, Qu, Leyuan, Li, Taihao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915751942684672
author Li, Haoxun
Sun, Yuqing
Shi, Hanlei
Liu, Yu
Qu, Leyuan
Li, Taihao
author_facet Li, Haoxun
Sun, Yuqing
Shi, Hanlei
Liu, Yu
Qu, Leyuan
Li, Taihao
contents Continuous dimensional speech emotion recognition captures affective variation along valence, arousal, and dominance, providing finer-grained representations than categorical approaches. Yet most multimodal methods rely solely on global transcripts, leading to two limitations: (1) all words are treated equally, overlooking that emphasis on different parts of a sentence can shift emotional meaning; (2) only surface lexical content is represented, lacking higher-level interpretive cues. To overcome these issues, we propose MSF-SER (Multi-granularity Semantic Fusion for Speech Emotion Recognition), which augments acoustic features with three complementary levels of textual semantics--Local Emphasized Semantics (LES), Global Semantics (GS), and Extended Semantics (ES). These are integrated via an intra-modal gated fusion and a cross-modal FiLM-modulated lightweight Mixture-of-Experts (FM-MOE). Experiments on MSP-Podcast and IEMOCAP show that MSF-SER consistently improves dimensional prediction, demonstrating the effectiveness of enriched semantic fusion for SER.
format Preprint
id arxiv_https___arxiv_org_abs_2510_05749
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MSF-SER: Enriching Acoustic Modeling with Multi-Granularity Semantics for Speech Emotion Recognition
Li, Haoxun
Sun, Yuqing
Shi, Hanlei
Liu, Yu
Qu, Leyuan
Li, Taihao
Sound
Continuous dimensional speech emotion recognition captures affective variation along valence, arousal, and dominance, providing finer-grained representations than categorical approaches. Yet most multimodal methods rely solely on global transcripts, leading to two limitations: (1) all words are treated equally, overlooking that emphasis on different parts of a sentence can shift emotional meaning; (2) only surface lexical content is represented, lacking higher-level interpretive cues. To overcome these issues, we propose MSF-SER (Multi-granularity Semantic Fusion for Speech Emotion Recognition), which augments acoustic features with three complementary levels of textual semantics--Local Emphasized Semantics (LES), Global Semantics (GS), and Extended Semantics (ES). These are integrated via an intra-modal gated fusion and a cross-modal FiLM-modulated lightweight Mixture-of-Experts (FM-MOE). Experiments on MSP-Podcast and IEMOCAP show that MSF-SER consistently improves dimensional prediction, demonstrating the effectiveness of enriched semantic fusion for SER.
title MSF-SER: Enriching Acoustic Modeling with Multi-Granularity Semantics for Speech Emotion Recognition
topic Sound
url https://arxiv.org/abs/2510.05749