aiqbench analysis
ProSarc introduces an audio-only framework that detects sarcasm by quantifying the mismatch between local prosodic dynamics and utterance-level emotional baselines. The model demonstrates strong generalization across spontaneous and cross-lingual datasets while providing interpretable uncertainty estimates that align with human perceptual ambiguity. This approach offers a robust alternative to frame-level labeling for temporal localization of sarcastic onsets in real-world speech scenarios.
ProSarc: Prosody-Aware Sarcasm Recognition Framework via Temporal Prosodic Incongruity
arXiv:2606.06168v1 Announce Type: new Abstract: We present ProSarc, an audio-only framework that detects sarcasm by modelling temporal prosodic incongruity, that is, the mismatch between local prosodic dynamics and the utterance-level emotional baseline. Dual encoding paths, a Global Emotion Encoder and a Temporal Prosody Encoder (BiLSTM + multi-head attention), feed a Prosodic Incongruity Analyzer that produces a scalar incongruity score for classification. Monte Carlo dropout provides uncertainty estimates, and an attention-based mechanism localises sarcastic onset without frame-level labels. ProSarc outperforms prior audio-only methods on MUStARD++ (F1=75.3) and generalises to spontaneous (PodSarc, F1=62.9) and cross-lingual speech (MuSaG, F1=65.6). Ten-run validation confirms the contribution of incongruity modelling (Wilcoxon p=0.002, Cohen's d=1.51). Human evaluation shows that model uncertainty tracks perceptual ambiguity and predicted onsets align with human-annotated temporal windows.
Abstract:We present ProSarc, an audio-only framework that detects sarcasm by modelling temporal prosodic incongruity, that is, the mismatch between local prosodic dynamics and the utterance-level emotional baseline. Dual encoding paths, a Global Emotion Encoder and a Temporal Prosody Encoder (BiLSTM + multi-head attention), feed a Prosodic Incongruity Analyzer that produces a scalar incongruity score for classification. Monte Carlo dropout provides uncertainty estimates, and an attention-based mechanism localises sarcastic onset without frame-level labels. ProSarc outperforms prior audio-only methods on MUStARD++ (F1=75.3) and generalises to spontaneous (PodSarc, F1=62.9) and cross-lingual speech (MuSaG, F1=65.6). Ten-run validation confirms the contribution of incongruity modelling (Wilcoxon p=0.002, Cohen's d=1.51). Human evaluation shows that model uncertainty tracks perceptual ambiguity and predicted onsets align with human-annotated temporal windows.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2606.06168 [cs.AI]
(or arXiv:2606.06168v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.06168 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 218 words
Related coverage
Modeling Sarcastic Speech: Semantic and Prosodic Cues in a Speech Synthesis Framework
arXiv · Jun 17, 2026
Computational Social Linguistics for Telugu Cultural Preservation: Novel Algorithms for Chandassu Metrical Pattern Recognition
arXiv · Oct 14, 2026
SCIC: Scope- and Codebook-Aware Instruction Conditioning for Speaker-Adapted Expressive TTS
arXiv · Oct 1, 2026
Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation
arXiv · Aug 20, 2026