DEAF: A Benchmark for Diagnostic Evaluation of Acoustic Faithfulness in Audio Language Models

• URL: http://arxiv.org/abs/2603.18048v2
• Date: Fri, 20 Mar 2026 01:17:57 GMT
• Title: DEAF: A Benchmark for Diagnostic Evaluation of Acoustic Faithfulness in Audio Language Models
• Abstract summary: Recent Audio Multimodal Large Language Models (Audio MLLMs) demonstrate impressive performance on speech benchmarks.<n>It remains unclear whether these models genuinely process acoustic signals or rely on text-based semantic inference.<n>We introduce DEAF, a benchmark of over 2,700 conflict stimuli spanning three acoustic dimensions.
• Score: 15.637236155997
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent Audio Multimodal Large Language Models (Audio MLLMs) demonstrate impressive performance on speech benchmarks, yet it remains unclear whether these models genuinely process acoustic signals or rely on text-based semantic inference. To systematically study this question, we introduce DEAF (Diagnostic Evaluation of Acoustic Faithfulness), a benchmark of over 2,700 conflict stimuli spanning three acoustic dimensions: emotional prosody, background sounds, and speaker identity. Then, we design a controlled multi-level evaluation framework that progressively increases textual influence, ranging from semantic conflicts in the content to misleading prompts and their combination, allowing us to disentangle content-driven bias from prompt-induced sycophancy. We further introduce diagnostic metrics to quantify model reliance on textual cues over acoustic signals. Our evaluation of seven Audio MLLMs reveals a consistent pattern of text dominance: models are sensitive to acoustic variations, yet predictions are predominantly driven by textual inputs, revealing a gap between high performance on standard speech benchmarks and genuine acoustic understanding.

Related papers

• Improving Text-to-Audio Instruction Following via Fine-Grained Feedback from Audio-Aware Large Language Models [51.75139461584678]
  Recent text-to-audio models generate high-quality audio, but often fail to follow instructions involving multiple sound events and temporal order.<n>We propose an instruction-level framework that uses audio-aware large language models as fine-grained judges to verify target event presence and temporal relations.<n>Experiments show that our method improves event completeness, temporal ordering, and joint instruction-following accuracy across existing benchmarks and S3Bench, while maintaining audio quality.
  arXiv  Detail & Related papers  (2026-07-15T03:13:06Z)
• Walking Through Uncertainty: An Empirical Study of Uncertainty Estimation for Audio-Aware Large Language Models [54.041320081289996]
  We present the first systematic empirical study of uncertainty estimation in audio-aware large language models (ALLMs)<n>We benchmark five representative methods, including predictive entropy, length-normalized entropy, semantic entropy, discrete semantic entropy, and P(True)<n>Our results reveal two key findings. First, semantic-level and verification-based methods consistently outperform token-level baselines on general audio reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-04-28T12:56:22Z)
• Evaluating Semantic Fragility in Text-to-Audio Generation Systems Under Controlled Prompt Perturbations [2.2870073664564115]
  Small linguistic changes may lead to substantial variation in generated audio, raising concerns about reliability in practical use.<n>We evaluate the semantic fragility of text-to-audio systems under controlled prompt perturbations.
  arXiv  Detail & Related papers  (2026-03-14T08:12:40Z)
• Evaluation of Audio Language Models for Fairness, Safety, and Security [10.580283172294733]
  We introduce a structural taxonomy (system-level and representational) of audio large language models (ALLMs)<n>We propose a unified evaluation framework that assesses semantic invariance under paralinguistic variation, refusal and toxicity behavior under unsafe prompts, and robustness to adversarial audio perturbations.<n>Our findings demonstrate that FSS behavior is tightly coupled to how acoustic information is integrated into semantic reasoning, underscoring the need for structure-aware evaluation of audio language models.
  arXiv  Detail & Related papers  (2026-02-25T21:22:28Z)
• Investigation for Relative Voice Impression Estimation [16.024524623990484]
  This study investigates relative voice impression estimation (RIE)<n>RIE is a framework for predicting the perceptual difference between two utterances from the same speaker.<n>We compare three modeling approaches: classical acoustic features commonly used for speech emotion recognition, self-supervised speech representations, and multimodal large language models.
  arXiv  Detail & Related papers  (2026-02-15T14:54:52Z)
• SpeakerSleuth: Evaluating Large Audio-Language Models as Judges for Multi-turn Speaker Consistency [12.420484491347073]
  We present SpeakerSleuth, a benchmark evaluating whether LALMs can reliably judge speaker consistency in multi-turn dialogues.<n>We construct 1,818 human-verified evaluation instances across four diverse datasets spanning synthetic and real speech.<n>We find that models struggle to reliably detect acoustic inconsistencies.
  arXiv  Detail & Related papers  (2026-01-07T15:45:41Z)
• AHELM: A Holistic Evaluation of Audio-Language Models [78.20477815156484]
  multimodal audio-language models (ALMs) take interleaved audio and text as input and output text.<n>AHELM is a benchmark that aggregates various datasets -- including 2 new synthetic audio-text datasets called PARADE and CoRe-Bench.<n>We also standardize the prompts, inference parameters, and evaluation metrics to ensure equitable comparisons across models.
  arXiv  Detail & Related papers  (2025-08-29T07:40:39Z)
• SpeechR: A Benchmark for Speech Reasoning in Large Audio-Language Models [60.72029578488467]
  SpeechR is a unified benchmark for evaluating reasoning over speech in large audio-language models.<n>It evaluates models along three key dimensions: factual retrieval, procedural inference, and normative judgment.<n> Evaluations on eleven state-of-the-art LALMs reveal that high transcription accuracy does not translate into strong reasoning capabilities.
  arXiv  Detail & Related papers  (2025-08-04T03:28:04Z)
• AudioJudge: Understanding What Works in Large Audio Model Based Speech Evaluation [55.607230723223346]
  This work presents a systematic study of Large Audio Model (LAM) as a Judge, AudioJudge, investigating whether it can provide a unified evaluation framework that addresses both challenges.<n>We explore AudioJudge across audio characteristic detection tasks, including pronunciation, speaking rate, speaker identification and speech quality, and system-level human preference simulation for automated benchmarking.<n>We introduce a multi-aspect ensemble AudioJudge to enable general-purpose multi-aspect audio evaluation. This method decomposes speech assessment into specialized judges for lexical content, speech quality, and paralinguistic features, achieving up to 0.91 Spearman correlation with human preferences on
  arXiv  Detail & Related papers  (2025-07-17T00:39:18Z)
• Challenge on Sound Scene Synthesis: Evaluating Text-to-Audio Generation [8.170174172545831]
  This paper addresses issues through the Sound Scene Synthesis challenge held as part of the Detection and Classification of Acoustic Scenes and Events 2024. We present an evaluation protocol combining objective metric, namely Fr'echet Audio Distance, with perceptual assessments, utilizing a structured prompt format to enable diverse captions and effective evaluation.
  arXiv  Detail & Related papers  (2024-10-23T06:35:41Z)
• AIR-Bench: Benchmarking Large Audio-Language Models via Generative Comprehension [95.8442896569132]
  We introduce AIR-Bench, the first benchmark to evaluate the ability of Large Audio-Language Models (LALMs) to understand various types of audio signals and interact with humans in the textual format. Results demonstrate a high level of consistency between GPT-4-based evaluation and human evaluation.
  arXiv  Detail & Related papers  (2024-02-12T15:41:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.