A scalable framework that identifies text-biased surface interpretations and converts disagreement regions into conflict QA examples and develops an agentic-style reasoning framework that converts speech into an Audio Twin, a text-readable representation of localized acoustic cues that exposes acoustic evidence to the reasoning model.
Abstract
Understanding spoken dialogue requires joint reasoning over lexical content and paralinguistic acoustic signals such as emotion and conversational intent. However, existing evaluations often allow shortcuts based on transcripts or single-modality solutions, obscuring whether models genuinely ground predictions in speech. We formalize this failure mode as cross-modal disagreement, where transcripts suggest plausible but incorrect surface interpretations while acoustic cues such as prosody or speaking style support different answers. We develop a scalable framework that identifies text-biased surface interpretations and converts disagreement regions into conflict QA examples. We also include consistent cases where transcript-based and speech-grounded interpretations agree, enabling evaluation beyond adversarial audio dependence. This results in ContraTalk, a controlled benchmark containing 501 questions across five discourse dimensions: interaction behavior, emotion state, dialogue act, social stance, and conversational intent. We further develop an agentic-style reasoning framework that converts speech into an Audio Twin, a text-readable representation of localized acoustic cues that exposes acoustic evidence to the reasoning model. Experiments show that strong text-only LLMs exceed 90% accuracy in consistent cases but drop to 33-48% in conflict cases. Direct AudioLLMs provide only partial grounding, still selecting the transcript-biased trap in roughly 30-40% of conflict cases. Our Audio Twin framework improves conflict-case accuracy while reducing trap selection, but its consistent-case behavior remains backbone-dependent. These results identify transcript-based shortcuts as an important failure mode in spoken dialogue understanding and show that explicit acoustic evidence aggregation provides a more controllable interface for diagnosing and improving speech-grounded reasoning.
Automated Text-to-Speech (TTS) evaluation methods (Mean Opinion Score (MOS) predictors and Audio Large Language Models (Audio-LLM) judges) are expected to reflect human perception, yet it is unclear how well they capture the distinct aspects of speech that listeners actually perceive. We deconstruct"naturalness"into a linguistically grounded annotation schema spanning 10 distinct perceptual dimensions, and use it to construct the first dimension-level meta-evaluation benchmark for TTS, comprising 860 utterances annotated by trained linguist raters. Results from benchmarking four MOS predictors and four Audio-LLM judges reveal that MOS predictors collapse onto acoustic signal quality, while Audio-LLM judges show selective, prompt-dependent detection that does not generalise across all dimensions. Neither class reliably captures a breadth of linguistically structured speech errors. Our dataset, annotation schema, and evaluation code are publicly released to support more targeted and interpretable TTS evaluation.
Oluwanifemi Bamgbose, Simon Rosen, J. Shah et al.· 0 citations
Negation is typically modeled through its linguistic realization, although spoken interaction is accompanied by tightly coordinated nonverbal behavior. We ask whether contexts centered on spoken negation cues contain measurable multimodal behavioral information: whether they can be distinguished from matched control contexts without lexical or acoustic input, where this information occurs in time, which modalities carry it, and whether it extends to the dialogue partner. We study 27 human-human interviews conducted in virtual reality, comprising temporally aligned gaze, facial, head, body, hand, and finger behavior and 964 annotated negation cues. Treating classification as a predictive probe, we compare 20 time-series models while excluding lexical and acoustic information, and then systematically vary temporal context, interactional source, modality availability, and event timing. Across grouped 10-fold cross-validation, the strongest probes reach up to .75 mean held-out AUROC from speaker-side behavior. Temporal analyses show that predictive information is concentrated around cue onset but remains detectable over a broader surrounding interval, while dialogue-partner behavior carries weaker predictive information with a comparatively diffuse temporal profile. Ablation and timing perturbations further show that facial features produce the largest modality-ablation effect and that the trained probe is sensitive to the temporal organization of the observed events.
Leon Hammerla, Patrick Schrottenbacher, Alexander Mehler· 0 citations
This work introduces TAKE 85, the first benchmark for directorial-intent understanding, comprising 398 short films with expert-verified question-answer pairs spanning global and fine-grained visual and audio intent, and establishes directorial intent as a previously overlooked dimension of multimodal understanding.
Kaishuu Shinozaki-Conefrey, Olivier Pascaud, Robin Courant et al.· 0 citations
A benchmark for evaluating whether models can infer the implicit, non-linear, and rhetorically layered meanings of social media videos that appear nonsensical on the surface but convey deliberate pragmatic meanings, and a diagnostic setting for measuring the gap between multimodal perception and pragmatic comprehension.
Yang Wang, Ya-Nan Ma, Yiqi Liu et al.· 0 citations
Emotion Recognition in Conversations (ERC) aims to identify speakers'emotions in multi-turn dialogue. Accurate emotion recognition can support a wide range of applications, including empathetic conversational agents, mental health support, and educational technologies. While many recent approaches rely on task-specific fine-tuning, such models may exploit dataset-specific cues. A central yet rarely questioned assumption in ERC is that each utterance can be assigned a single unambiguous emotion label. To investigate this assumption, we study ERC using Large Language Models (LLMs) in a zero-shot setting while incorporating preceding conversational turns as context. We show that aggregate metrics mask systematic failures. Errors concentrate around utterances containing negations, exclamations, and interjections. This pattern is consistent across all evaluated models, suggesting limitations in the benchmarks rather than model-specific weaknesses. A controlled re-annotation study involving four human annotators supports this finding: strong agreement is observed in only 35 percent of cases, with neutral utterances dominating high-agreement instances, while many emotional categories fall into low-agreement regimes. These findings suggest that many apparent model errors reflect genuine annotation ambiguity rather than poor emotion understanding. Standard single-label evaluation is therefore insufficient. To address this limitation, we introduce an LLM-as-Judge framework that evaluates each emotion independently according to its plausibility in the conversational context rather than enforcing a single-label decision.
Amir Ben Khalifa, Fanny Bezancon, Bessam Abdulrazak et al.· 0 citations
Hear2Act is introduced, a unified evaluation protocol for text and spoken assistants with 480 persona-grounded scenarios, hidden user concerns, and objectively verifiable outcomes that show that prosody matters when lexical evidence is insufficient, and that audio-capable LLMs can recover information from speech but do not reliably carry it into action without an explicit intermediate representation.
Xinyi Liu, H. Nayyeri, Dilek Hakkani-Tur et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
