VoxEmo: Benchmarking Speech Emotion Recognition with Speech LLMs
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917327819243520 |
|---|---|
| author | Zhang, Hezhao Chou, Huang-Cheng Narayanan, Shrikanth Hain, Thomas |
| author_facet | Zhang, Hezhao Chou, Huang-Cheng Narayanan, Shrikanth Hain, Thomas |
| contents | Speech Large Language Models (LLMs) show great promise for speech emotion recognition (SER) via generative interfaces. However, shifting from closed-set classification to open text generation introduces zero-shot stochasticity, making evaluation highly sensitive to prompts. Additionally, conventional speech LLMs benchmarks overlook the inherent ambiguity of human emotion. Hence, we present VoxEmo, a comprehensive SER benchmark encompassing 35 emotion corpora across 15 languages for Speech LLMs. VoxEmo provides a standardized toolkit featuring varying prompt complexities, from direct classification to paralinguistic reasoning. To reflect real-world perception/application, we introduce a distribution-aware soft-label protocol and a prompt-ensemble strategy that emulates annotator disagreement. Experiments reveal that while zero-shot speech LLMs trail supervised baselines in hard-label accuracy, they uniquely align with human subjective distributions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08936 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VoxEmo: Benchmarking Speech Emotion Recognition with Speech LLMs Zhang, Hezhao Chou, Huang-Cheng Narayanan, Shrikanth Hain, Thomas Sound Artificial Intelligence Computation and Language Multimedia Audio and Speech Processing Speech Large Language Models (LLMs) show great promise for speech emotion recognition (SER) via generative interfaces. However, shifting from closed-set classification to open text generation introduces zero-shot stochasticity, making evaluation highly sensitive to prompts. Additionally, conventional speech LLMs benchmarks overlook the inherent ambiguity of human emotion. Hence, we present VoxEmo, a comprehensive SER benchmark encompassing 35 emotion corpora across 15 languages for Speech LLMs. VoxEmo provides a standardized toolkit featuring varying prompt complexities, from direct classification to paralinguistic reasoning. To reflect real-world perception/application, we introduce a distribution-aware soft-label protocol and a prompt-ensemble strategy that emulates annotator disagreement. Experiments reveal that while zero-shot speech LLMs trail supervised baselines in hard-label accuracy, they uniquely align with human subjective distributions. |
| title | VoxEmo: Benchmarking Speech Emotion Recognition with Speech LLMs |
| topic | Sound Artificial Intelligence Computation and Language Multimedia Audio and Speech Processing |
| url | https://arxiv.org/abs/2603.08936 |