The Model Hears You: Audio Language Model Deployments Should Consider the Principle of Least Privilege

• URL: http://arxiv.org/abs/2503.16833v2
• Date: Tue, 09 Sep 2025 00:51:12 GMT
• Title: The Model Hears You: Audio Language Model Deployments Should Consider the Principle of Least Privilege
• Authors: Luxi He, Xiangyu Qi, Michel Liao, Inyoung Cheong, Prateek Mittal, Danqi Chen, Peter Henderson,
• Abstract summary: Latest Audio Language Models (Audio LMs) process speech directly instead of relying on a separate transcription step.<n>This shift preserves detailed information, such as intonation or the presence of multiple speakers, that would otherwise be lost in transcription.<n>It also introduces new safety risks, including the potential misuse of speaker identity cues and other sensitive vocal attributes.
• Score: 48.18013944679755
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The latest Audio Language Models (Audio LMs) process speech directly instead of relying on a separate transcription step. This shift preserves detailed information, such as intonation or the presence of multiple speakers, that would otherwise be lost in transcription. However, it also introduces new safety risks, including the potential misuse of speaker identity cues and other sensitive vocal attributes, which could have legal implications. In this paper, we urge a closer examination of how these models are built and deployed. Our experiments show that end-to-end modeling, compared with cascaded pipelines, creates socio-technical safety risks such as identity inference, biased decision-making, and emotion detection. This raises concerns about whether Audio LMs store voiceprints and function in ways that create uncertainty under existing legal regimes. We then argue that the Principle of Least Privilege should be considered to guide the development and deployment of these models. Specifically, evaluations should assess (1) the privacy and safety risks associated with end-to-end modeling; and (2) the appropriate scope of information access. Finally, we highlight related gaps in current audio LM benchmarks and identify key open research questions, both technical and policy-related, that must be addressed to enable the responsible deployment of end-to-end Audio LMs.

Related papers

• Now You Hear Me: Audio Narrative Attacks Against Large Audio-Language Models [26.648297855855432]
  We design a text-to-audio jailbreak that embeds disallowed directives within a narrative-style audio stream.<n>The attack exploits structural and acoustic properties, thereby circumventing safety mechanisms primarily calibrated for text.<n>Results highlight the need for safety frameworks that jointly reason over linguistic and paralinguistic representations.
  arXiv  Detail & Related papers  (2026-01-30T18:23:02Z)
• WESR: Scaling and Evaluating Word-level Event-Speech Recognition [59.21814194620928]
  Speech conveys not only linguistic information but also rich non-verbal vocal events such as laughing and crying.<n>We develop a refined taxonomy of 21 vocal events, with a new categorization into discrete (standalone) versus continuous (mixed with speech) types.<n>Based on the refined taxonomy, we introduce WESR-Bench, an expert-annotated evaluation set (900+ utterances) with a novel position-aware protocol.
  arXiv  Detail & Related papers  (2026-01-08T02:23:21Z)
• 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)
• Evaluating Language Model Reasoning about Confidential Information [95.64687778185703]
  We study whether language models exhibit contextual robustness, or the capability to adhere to context-dependent safety specifications.<n>We develop a benchmark (PasswordEval) that measures whether language models can correctly determine when a user request is authorized.<n>We find that current open- and closed-source models struggle with this seemingly simple task, and that, perhaps surprisingly, reasoning capabilities do not generally improve performance.
  arXiv  Detail & Related papers  (2025-08-27T15:39:46Z)
• From Alignment to Advancement: Bootstrapping Audio-Language Alignment with Synthetic Data [55.2480439325792]
  Audio-aware large language models (ALLMs) have recently made great strides in understanding and processing audio inputs.<n>These models are typically adapted from text-based large language models (LLMs) through additional training on audio-related tasks.<n>We propose a data generation framework that produces contrastive-like training data, designed to enhance ALLMs' ability to differentiate between present and absent sounds.
  arXiv  Detail & Related papers  (2025-05-26T16:08:41Z)
• Baichuan-Audio: A Unified Framework for End-to-End Speech Interaction [9.101978573666546]
  Baichuan-Audio is an end-to-end audio large language model that seamlessly integrates audio understanding and generation.<n>It features a text-guided aligned speech generation mechanism, enabling real-time speech interaction with both comprehension and generation capabilities.
  arXiv  Detail & Related papers  (2025-02-24T15:16:34Z)
• VoxEval: Benchmarking the Knowledge Understanding Capabilities of End-to-End Spoken Language Models [32.086847480051084]
  We present VoxEval, a novel SpeechQA benchmark that assesses knowledge understanding through pure speech interactions.<n>Our benchmark 1) maintains speech format for both inputs and outputs, 2) evaluates model robustness across diverse input audio conditions, and 3) pioneers the assessment of complex tasks like mathematical reasoning in spoken format.
  arXiv  Detail & Related papers  (2025-01-09T04:30:12Z)
• MSA-ASR: Efficient Multilingual Speaker Attribution with frozen ASR Models [59.80042864360884]
  Speaker-attributed automatic speech recognition (SA-ASR) aims to transcribe speech while assigning transcripts to the corresponding speakers accurately. This paper introduces a novel approach, leveraging a frozen multilingual ASR model to incorporate speaker attribution into the transcriptions.
  arXiv  Detail & Related papers  (2024-11-27T09:01:08Z)
• Recent Advances in Speech Language Models: A Survey [45.968078636811356]
  Speech Language Models (SpeechLMs) are end-to-end models that generate speech without converting from text.<n>This survey paper provides the first comprehensive overview of recent methodologies for constructing SpeechLMs.
  arXiv  Detail & Related papers  (2024-10-01T21:48:12Z)
• Enhancing Temporal Understanding in Audio Question Answering for Large Audio Language Models [0.9285295512807729]
  The Audio Question Answering (AQA) task includes audio event classification, audio captioning, and open-ended reasoning.<n>LALMs excel in general audio understanding, but are limited in temporal reasoning.<n>This paper addresses these challenges and limitations in audio temporal reasoning.
  arXiv  Detail & Related papers  (2024-09-10T05:26:53Z)
• SpeechPrompt: Prompting Speech Language Models for Speech Processing Tasks [94.10497337235083]
  We are first to explore the potential of prompting speech LMs in the domain of speech processing. We reformulate speech processing tasks into speech-to-unit generation tasks. We show that the prompting method can achieve competitive performance compared to the strong fine-tuning method.
  arXiv  Detail & Related papers  (2024-08-23T13:00:10Z)
• 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)
• It's Never Too Late: Fusing Acoustic Information into Large Language Models for Automatic Speech Recognition [70.77292069313154]
  Large language models (LLMs) can be successfully used for generative error correction (GER) on top of the automatic speech recognition (ASR) output. In this work, we aim to overcome such a limitation by infusing acoustic information before generating the predicted transcription through a novel late fusion solution termed Uncertainty-Aware Dynamic Fusion (UADF)
  arXiv  Detail & Related papers  (2024-02-08T07:21:45Z)
• Learning Speech Representation From Contrastive Token-Acoustic Pretraining [57.08426714676043]
  We propose "Contrastive Token-Acoustic Pretraining (CTAP)", which uses two encoders to bring phoneme and speech into a joint multimodal space. The proposed CTAP model is trained on 210k speech and phoneme pairs, achieving minimally-supervised TTS, VC, and ASR.
  arXiv  Detail & Related papers  (2023-09-01T12:35:43Z)
• Why So Gullible? Enhancing the Robustness of Retrieval-Augmented Models against Counterfactual Noise [14.38859858538404]
  In a retrieved document set, even the "relevant" documents may contain misleading or incorrect information. Our work investigates a more challenging scenario in which even the "relevant" documents may contain misleading or incorrect information. We propose approaches for handling knowledge conflicts among retrieved documents by explicitly fine-tuning a discriminator or prompting GPT-3.5 to elicit its discriminative capability.
  arXiv  Detail & Related papers  (2023-05-02T16:28:10Z)
• Introducing Model Inversion Attacks on Automatic Speaker Recognition [0.9558392439655015]
  Model inversion (MI) attacks allow to reconstruct average per-class representations of a machine learning (ML) model's training data. We present an approach to (1) reconstruct audio samples from a trained ML model and (2) extract intermediate voice feature representations which provide valuable insights into the speakers' biometrics. Our sliding MI extends standard MI by iteratively inverting overlapping chunks of the audio samples. We show that one can use the inverted audio data to generate spoofed audio samples to impersonate a speaker, and execute voice-protected commands for highly secured systems.
  arXiv  Detail & Related papers  (2023-01-09T08:51:15Z)
• Cross-lingual Text-To-Speech with Flow-based Voice Conversion for Improved Pronunciation [11.336431583289382]
  This paper presents a method for end-to-end cross-lingual text-to-speech. It aims to preserve the target language's pronunciation regardless of the original speaker's language.
  arXiv  Detail & Related papers  (2022-10-31T12:44:53Z)
• Towards Language Modelling in the Speech Domain Using Sub-word Linguistic Units [56.52704348773307]
  We propose a novel LSTM-based generative speech LM based on linguistic units including syllables and phonemes. With a limited dataset, orders of magnitude smaller than that required by contemporary generative models, our model closely approximates babbling speech. We show the effect of training with auxiliary text LMs, multitask learning objectives, and auxiliary articulatory features.
  arXiv  Detail & Related papers  (2021-10-31T22:48:30Z)
• Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages [14.297371692669545]
  Unsupervised acoustic modeling (UAM) for automatic speech recognition (ASR) in the zero-resource scenario. First problem concerns unsupervised discovery of basic (subword level) speech units in a given language. Second problem is referred to as unsupervised subword modeling.
  arXiv  Detail & Related papers  (2020-07-29T19:45:17Z)

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.