Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Jingyi, Guo, Zhimeng, Chun, Jiyun, Wang, Pichao, Perrault, Andrew, Elsner, Micha
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2510.10444
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908598440820736
author Chen, Jingyi
Guo, Zhimeng
Chun, Jiyun
Wang, Pichao
Perrault, Andrew
Elsner, Micha
author_facet Chen, Jingyi
Guo, Zhimeng
Chun, Jiyun
Wang, Pichao
Perrault, Andrew
Elsner, Micha
contents Understanding emotion from speech requires sensitivity to both lexical and acoustic cues. However, it remains unclear whether large audio language models (LALMs) genuinely process acoustic information or rely primarily on lexical content. We present LISTEN (Lexical vs. Acoustic Speech Test for Emotion in Narratives), a controlled benchmark designed to disentangle lexical reliance from acoustic sensitivity in emotion understanding. Across evaluations of six state-of-the-art LALMs, we observe a consistent lexical dominance. Models predict "neutral" when lexical cues are neutral or absent, show limited gains under cue alignment, and fail to classify distinct emotions under cue conflict. In paralinguistic settings, performance approaches chance. These results indicate that current LALMs largely "transcribe" rather than "listen," relying heavily on lexical semantics while underutilizing acoustic cues. LISTEN offers a principled framework for assessing emotion understanding in multimodal models.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10444
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Do Audio LLMs Really LISTEN, or Just Transcribe? Measuring Lexical vs. Acoustic Emotion Cues Reliance
Chen, Jingyi
Guo, Zhimeng
Chun, Jiyun
Wang, Pichao
Perrault, Andrew
Elsner, Micha
Computation and Language
Artificial Intelligence
Understanding emotion from speech requires sensitivity to both lexical and acoustic cues. However, it remains unclear whether large audio language models (LALMs) genuinely process acoustic information or rely primarily on lexical content. We present LISTEN (Lexical vs. Acoustic Speech Test for Emotion in Narratives), a controlled benchmark designed to disentangle lexical reliance from acoustic sensitivity in emotion understanding. Across evaluations of six state-of-the-art LALMs, we observe a consistent lexical dominance. Models predict "neutral" when lexical cues are neutral or absent, show limited gains under cue alignment, and fail to classify distinct emotions under cue conflict. In paralinguistic settings, performance approaches chance. These results indicate that current LALMs largely "transcribe" rather than "listen," relying heavily on lexical semantics while underutilizing acoustic cues. LISTEN offers a principled framework for assessing emotion understanding in multimodal models.
title Do Audio LLMs Really LISTEN, or Just Transcribe? Measuring Lexical vs. Acoustic Emotion Cues Reliance
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.10444