The paper introduces an incongruence score that quantifies disagreement between a speaker’s verbal content and vocal expression in a shared valence‑arousal‑dominance space, validated against human judgments. Using this metric, the authors evaluate three audio‑capable large language models across acted, conversational, and spontaneous speech datasets, finding that higher incongruence consistently increases error rates when audio is present. The study also shows that supervised classifiers behave oppositely, suggesting the challenge lies in how models fuse multimodal evidence rather than in the signal itself, and that the score remains predictive even without an emotion lexicon.
arXiv:2609.39453v1 Announce Type: cross Abstract: Speech emotion recognition (SER) is the task of assigning emotion labels to utterances. Early systems relied on acoustic features, whereas recent app...
arXiv:2608.28932v1 Announce Type: new Abstract: Voice products increasingly need affective cues that are present in speech but absent from transcripts. We introduce VocalAffectBench, a public, test-o...
The paper introduces an LLM-based framework for continuous dimensional emotion evaluation in multimodal dialogue, combining discrete emotion recognition with Valence-Arousal-Dominance (VAD) assessment on the IEMOCAP dataset. It incorporates acoustic cues as natural language descriptions via the SpeechCueLLM approach and evaluates six models from the LLaMA, GPT, and Qwen families using zero-shot, few-shot, and LoRA fine-tuning. LoRA-fine-tuned LLaMA models outperform prompt-engineered GPT models, achieving a new state-of-the-art Valence CCC of 0.7822, and ablation studies show that textual audio descriptions significantly benefit smaller models. "whyItMatters":"The study demonstrates that domain adaptation through fine-tuning can surpass larger GPT models in multimodal emotion evaluation, highlighting the importance of tailored training for emotion recognition tasks."
arXiv:2606. 07309v1 Announce Type: cross Abstract: Instruction-following audio language models (ALMs) can be augmented with explicit acoustic cues, yet it remains unclear whether such cues are used in a grounded way when the raw audio is already available.
The paper compares text‑based and feature‑based models for recognizing compound emotions in real‑world videos. It proposes textualizing non‑verbal cues from audio and visual modalities into text to leverage large language models, while feature‑based models directly combine extracted multimodal features. Experiments on the C‑EXPR‑DB dataset show that feature‑based models outperform textualization in the wild, though textual models can excel when rich transcripts are available.
arXiv:2606. 24941v2 Announce Type: replace-cross Abstract: Reviewing recorded interviews for affective cues such as composure and agitation is slow and subjective, and cloud services that could automate the task require sensitive audio to leave the device.
arXiv:2606. 00670v1 Announce Type: cross Abstract: Face-to-face speech comprehension is inherently multimodal, integrating acoustic signals with visible articulation, facial expression, head motion, and other socially relevant cues.
The paper introduces a multimodal emotion recognition system that integrates facial expressions, speech, text, and body movement to aid clinicians in psychological evaluation. While individual modalities perform well on benchmarks, their real‑world accuracy drops to 82.97%–87.64%, whereas the fused multimodal system reaches 96.43% accuracy with no state below 92.31%. These findings demonstrate that multimodal fusion enhances robustness in realistic settings, supporting more reliable evaluator assistance.
arXiv:2608. 06409v1 Announce Type: cross Abstract: Speech language models are increasingly evaluated on paralinguistic tasks by the accuracy of prompted answers, but answer accuracy combines failures at different stages of the audio-to-answer computation.
arXiv:2606. 27717v1 Announce Type: cross Abstract: Prosodic emphasis varies across languages, emotions, and speaking styles, yet existing emphasis detection models are largely trained and evaluated on monolingual neutral read speech.
SpeechCritic learns a diagnostic speech judge from only about 300 human-labeled comparisons, calibrating a frontier audio-language model with non-binding hints that select acoustic measurements agreeing with human judgments. The approach improves dimension-level agreement by 6.3 points and reduces mismatch in Tie rates by 10.4 points, and training a 7B judge with SFT, OPD, and RL yields distinct behaviors, with RL producing more specific rationales. The method works across language pairs, demonstrating a path from limited human preferences to a diagnostic speech judge.
arXiv:2606. 15888v1 Announce Type: cross Abstract: Non-verbal vocalizations (NVs), such as laughter, sighs, and coughs, are important acoustic cues for emotion and intent.