Audio Language Models

Modern neural audio models run on accelerators whose peak throughput comes from dense matrix multiplication, increasingly at the edge and in datacenters. The conventional acoustic frontend, however -- a Short-Time Fourier Transform (STFT) followed by sparse Mel aggregation -- remains a multi-stage pipeline centered on the Fast Fourier Transform (FFT), with execution overheads unlike the dense linear algebra dominating the inference stack. This work introduces MelT, a portable single-stage Mel frontend that precomputes Mel-spaced Non-Uniform Discrete Fourier Transform (NDFT) bases and applies them to time-domain frames through General Matrix Multiplication (GEMM). The contribution is a computational design principle: decoupling Mel feature extraction from vendor-specific FFT primitives and lowering it onto the matrix-multiplication substrate accelerators already optimize. It is not a new spectral operator. MelT's direct projection performs more arithmetic than the FFT pipeline. Yet in the compact-resolution regime of neural audio frontends, it achieves a 1.64-times to 3.29-times latency reduction and up to a 3.03-times reduction in measured active energy, from the Apple A18 Pro to the NVIDIA H100. All gains are within-platform comparisons, accompanied by task-level validation. Word error rate stays statistically equivalent to the native frontend's on frozen Whisper models of medium size and larger; speaker-attribute classification on VoxCeleb1 is non-inferior. The cepstral extension MFCCT preserves utility on a clinical respiratory-insufficiency classification task (SPIRA) while improving on the MFCC baseline. These results indicate that, in practical regimes on the accelerators evaluated here, hardware alignment rather than arithmetic count can govern the realized cost of feature extraction.

Sound design workflows frequently oscillate between time-consuming library searches and the complexity of procedural synthesis, with practitioners typically relying on disconnected tools to address each challenge separately. This paper introduces Quality Audio Prototyping (QuAP), a working prototype that unifies content-based audio retrieval and procedural sound generation within a single interface, reducing the procedural distance between a narrative concept and its sonic realisation. QuAP integrates a similarity-based retrieval engine with real-time procedural audio models, complemented by a rule-based assistant that provides perceptually informed parameter guidance, offering definitions and recommendations derived from empirical optimisation rather than requiring prior synthesis knowledge. Preliminary evaluation confirms the viability of this approach: subjective assessment demonstrated statistically significant quality improvements in five of six embedded synthesis models, and an encoder ablation study established the preferred retrieval architecture on a sound effect dataset. A user evaluation with 16 practitioners confirmed the tool's workflow utility, with all participants agreeing that the parameter assistant preserved creative agency while lowering the barrier to procedural interaction.

We present UNISON, a latent diffusion framework that unifies speech generation, sound generation, and audio editing within a single model. A single model handles text-to-audio, text-to-speech, zero-shot speaker cloning, mixed speech-and-sound generation, scene-level audio editing, speech-in-scene editing, and timed temporal composition, all of which share a single set of weights. Our architecture features two core designs: (1) Layer-wise deep LLM fusion, which injects hidden states from uniformly sampled layers of a frozen MLLM into corresponding MM-DiT blocks via learned projections, providing depth-matched semantic conditioning that improves instruction following over single-layer baselines; and (2) a unified multi-task architecture where task identity is encoded solely by a channel-wise mask and source audio is provided through VAE-encoded channel concatenation. Training is stabilized by an online GPU-side multi-task data synthesis pipeline with task-homogeneous batching and a two-stage curriculum. With 621M--732M trainable parameters, UNISON achieves results competitive with or exceeding task-specialist models across evaluated domains, while being roughly $4\times$ smaller than comparable unified systems.

Semantic speech tokenizers have become a widely used interface for Audio-LLMs, owing to their compact single-codebook design and strong linguistic alignment. However, their focus on linguistic abstraction induces acoustic blindness, limiting their applicability beyond speech-centric tasks. We propose UniAudio-Token, a framework that empowers semantic tokenizers with general audio perception without compromising speech ability. Instead of altering the semantic paradigm, UniAudio-Token mitigates its information loss through two key innovations: (1) Semantic-Acoustic Primitives (SAP) provide structured supervision by decomposing audio into linguistic content, vocal attributes, and auditory-scene primitives; and (2) Semantic-Acoustic Equilibrium (SAE) introduces a content-aware gating mechanism that adaptively restores fine-grained acoustic details from shallow layers. Extensive evaluations show that UniAudio-Token learns comprehensive universal representations while preserving high-fidelity speech generation. When integrated with downstream LLMs, it outperforms all single-codebook baseline tokenizers on both understanding and generation tasks, effectively serving as a unified audio interface. We publicly release all our code, including training and inference scripts, together with the model checkpoints at https://github.com/Tencent/Universal_Audio_Tokenizer.

Controllable music editing is to modify high-level attributes while strictly preserving rhythmic and melodic structures. However, this task is challenged by a semantic-structural entanglement: steering methods often degrade structure to achieve editing performance, while structural adaptors suppress semantic responsiveness. We propose AnchorSteer, a framework that disentangles this tension by coupling structural anchoring with self-discovered semantic steering. The proposed approach probes internal representations to extract interpretable, label-free concept vectors via a self-supervised reconstruction objective, isolating attributes without curated data. During editing, these portable, plug-and-play concept vectors are injected into diffusion hidden manifolds while a structural adaptor enforces consistency. Variants for unconditioned and conditioned injections are provided to balance robustness and semantic strength. Experiments on ZoME-Bench and subjective tests show that the proposed framework outperforms both steering-only and anchoring-only baselines, enabling significant semantic transformations with high-fidelity structural preservation.

Real-time and accurate spatial audio generation is pivotal for delivering an immersive experience. However, existing spatial audio synthesis technologies are often encumbered by a tradeoff between generation quality and high inference latency, as well as difficulty in capturing precise spatial information from multimodal inputs. To address these challenges, we propose SwanSphere, a unified streaming framework for high-fidelity spatial audio generation from panoramic videos and text prompts. SwanSphere mainly makes the following contributions: 1) We introduce a causal autoregressive diffusion transformer architecture that enables streaming high-quality spatial audio generation. 2) We design a Spatial Video-Audio Contrastive (SVAC) learning strategy to align the video encoder with the acoustic domain, and further employ a multi-objective online direct preference optimization (ODPO) scheme, resulting in strong spatial perception and robust multimodal spatial audio synthesis. 3) To alleviate the current scarcity of spatial audio datasets, we also develop an automated annotation pipeline for generating detailed spatial captions. Experimental results demonstrate that SwanSphere achieves superior performance in both video-to-spatial and text-to-spatial audio generation tasks. Demos can be found at: https://swanaigc.github.io.

With the rise of AI-generated audio, watermarking has become widely used for detecting misuse and protecting intellectual property. However, adversaries may try to remove these watermarks, making it critical to evaluate how well watermarking schemes withstand removal attacks. Existing attacks are often impractical: they either noticeably degrade perceptual quality or require access to the watermarking scheme. We propose DiffErase, a black-box watermark removal attack that assumes no knowledge of the target watermarking scheme while maintaining perceptual quality. DiffErase perturbs watermarked audio to an intermediate diffusion noise level and regenerates it using a pretrained denoising model, effectively suppressing watermark signals. Theoretical analysis and extensive experiments demonstrate that inaudible audio watermarks are highly vulnerable: across multiple audio domains, DiffErase consistently removes watermarks while preserving perceptual quality. These findings highlight the need for future audio watermarking designs to consider diffusion-based threats. Code and demos are available at https://differase.github.io/DiffErase/.

3D audio and novel-view acoustic synthesis models are usually evaluated with global metrics.However, global metrics often hide where and why binaural prediction fails. We propose a full-reference diagnostic framework that uses time-frequency audio error maps for magnitude, ILD, IPD, temporal alignment, loudness, and high-frequency failures, forming a 3D Audio Error Map (3DAE Map) for visual inspection. We frame these diagnostics into a model-agnostic benchmark, Spatial Audio Error Bench (3DAE Bench), which takes arbitrary ground-truth and predicted binaural pairs and reports the prediction quality of audio novel-view synthesis models. Experiments on ViGAS outputs over Replay-NVAS and SoundSpaces show different dominant failure modes: temporal misalignment on Replay-NVAS and ILD mismatch on SoundSpaces. Overall, the framework provides interpretable failure-mode summaries and intuitive visual maps for audio Novel-view-synthesis model development optimization.

Large Audio Language Models (LALMs) expand jailbreak risks from token-level prompting to the full speech perception-to-reasoning pipeline, where unsafe behavior can be induced through semantics, acoustic style, signal artifacts, or internal representations. Existing work studies these risks under heterogeneous threat models and evaluation protocols, making it difficult to compare attack practicality or defense utility. This paper provides a unified taxonomy and a controlled empirical evaluation of LALM jailbreak attacks and defenses. We organize prior work into semantic, acoustic, signal, and embedding-layer attacks; guard-based, training-free, and training-based defenses; and cross-modal, audio-native, and interactive benchmarks. We then evaluate representative attacks and defenses across ten open-source LALMs, measuring not only attack success rate but also benign refusal and latency. Our results show that Acoustic Best-of-N reveals strong worst-case audio-space vulnerabilities, Narrative Framing is an effective low-latency semantic threat, and current defenses trade robustness against benign usability. These findings support cost- and utility-aware evaluation as a necessary complement to success-rate-only LALM safety benchmarks.

Contrastive Language-Audio Pretraining (CLAP) models are widely used for audio understanding and support modality-agnostic condition swapping in many zero-shot applications. However, their performance is heavily affected by the modality gap between audio and text embeddings. Existing explanations mainly attribute this gap to the cone effect, treating it as a shift between mean embeddings, yet correcting the mean alone yields only limited improvements. Alternative hypotheses, such as information imbalance and dimensionality collapse, have also been proposed, but they remain insufficiently verified and have not been thoroughly studied in the audio domain. Meanwhile, several works attempt to decompose multimodal contrastive embeddings into interpretable concepts, but none explicitly analyze the modality gap from the perspective of concept decomposition. In this work, we introduce COMET (Concept space Organization and Modality gap Explanation with PLS-SVD Transformation), a novel partial least squares singular value decomposition (PLS-SVD) framework for CLAP that unveils a broader perspective of the modality gap. Our framework reveals that only a small, interpretable subset of axes, which captures shared concepts, contributes substantially to similarity computation, and that the mean component represents only partially the modality gap. Building on this insight, we propose a simple spectral truncation method that mitigates the modality gap in a training-free manner. The method enables zero-shot audio captioning with condition swapping to approach fully supervised performance, without requiring large auxiliary memory banks or expensive computation. At the same time, it achieves substantial embedding dimensionality reduction while preserving strong performance on retrieval and audio captioning tasks.

Recent Large Audio-Language Models (LALMs) have demonstrated promising abilities in understanding musical content. However, whether their responses are grounded in the correct temporal regions of the audio remains underexplored. This limitation is particularly critical for music understanding, where key information often occurs as temporally localized events, such as instrument entries and rhythmic transitions. To address this gap, we introduce MusTBench, a music-expert-validated benchmark designed to evaluate temporal grounding in LALMs through five temporally grounded question-answering tasks. To further improve temporal grounding in existing models, we propose MusT, a novel four-stage temporal optimization recipe spanning music encoder adaptation, LLM adaptation, LLM supervised fine-tuning, and RL-based optimization. Experiments on MusTBench show that existing LALMs struggle with precise temporal grounding, while MusT brings significant improvements over strong baselines. These results establish temporal grounding as a key missing capability in current LALMs and position MusTBench as a challenging benchmark for future research in temporally grounded music understanding.

Recent advances in text-to-music generation enable high-fidelity synthesis of structured musical audio, raising growing concerns about data provenance, consent, and training transparency. These models are typically trained on large-scale corpora with little disclosure, leaving no practical mechanism to verify whether a particular audio sample was included in training. In this paper, we investigate black-box membership inference for generative music models, aiming to determine whether a candidate music sample was used during training, given only query access to the deployed system. Our key insight is that training membership induces systematically stronger semantic and structural alignment between a candidate sample and the model's generation conditioned on its caption. We query the target model with the associated caption and measure the relationship between the candidate audio and the generated output in a learned feature space. To capture features that separate members from non-members, we construct paired examples consisting of each track and its caption-conditioned generation from shadow models, and train a music auditor to classify membership. The auditor captures alignment patterns characteristic of training membership and generalizes to unseen target models in a fully black-box setting without access to model parameters or training metadata. Across multiple state-of-the-art music generators, our method achieves up to 98.6% accuracy, with false-positive and false-negative rates as low as 1.9% and 1.0%, demonstrating that reliable training-data auditing is feasible in realistic deployment scenarios.

The use of Large Language Models (LLMs) is proliferating, yet their performance is observed to vary based on prompting styles and tones. In this study, we investigate both whether and how tonal variations in prompts lead to disparate LLM accuracy for objective multiple-choice questions. We use two datasets: a 50-base question dataset with five tone variants and a 570-base question MMLU subset spanning 57 subjects with seven tone variants. Experiments were conducted to evaluate the performance of four cost-efficient, popular LLMs: ChatGPT-4o, ChatGPT-5-nano, Gemini 2.5 Flash, and Gemini 2.5 Flash Lite. Across models, tonal effects are systematic but highly model-dependent. Some models show small, yet statistically significant, shifts, while others exhibit large accuracy swings across tones. Further, we identify subject-level differences in tone sensitivity and present a routing framework to explain how tones may attune internal reasoning modes. Our findings caution users against assuming tone-robust reliability in LLM deployments.

We present DEMON, a real-time diffusion engine that makes the denoising process playable as a live musical instrument: a control surface both broad (many parameters shaped per-frame across the output) and responsive (each control taking effect as fast as its place in the denoising loop allows). Built on ACE-Step 1.5 and StreamDiffusion's ring-buffer architecture with TensorRT acceleration, it sustains up to 12.3 decoder completions per second for 60-second music on a single consumer GPU (RTX 5090), or 11.3 generations per second at our production ring-depth of 4. At these rates denoising parameters become viable as live performance controls, but the ring buffer propagates per-request changes only at its drain rate, a floor of S denoising steps. We contribute four mechanisms. (1) Per-slot heterogeneous denoise scheduling: each ring-buffer slot owns its timestep schedule, so a moving denoise slider is tracked without wiping the in-flight queue, where the upstream global-schedule design must rebuild and discard it. (2) Shared mutable per-step state, giving any parameter consulted at every solver step next-tick effect, bypassing ring-buffer drain. (3) Per-frame source blending: a sampling-time control on the standard SDE re-noise step, giving a framewise transformation-strength axis that complements scalar denoise scheduling. (4) Windowed VAE decode exploiting receptive-field analysis for an 8.0x decode speedup. Together these separate streaming-diffusion parameters into four propagation classes, by onset and convergence latency.

Audio agents extend large audio-language models (LALMs) by decomposing audio questions into tool calls, intermediate evidence, and iterative reasoning steps. However, as LALMs become stronger, the key challenge shifts from enabling tool use to determining when agentic evidence acquisition genuinely benefits audio understanding. We propose Audio-Mind, an auditable and pluggable framework for conditional evidence acquisition in audio understanding. Audio-Mind dynamically combines a strong frontend with planner-guided tool use, preserving frontend judgment when initial evidence is sufficient while acquiring bounded external evidence for questions with unresolved evidence gaps. Experiments on MMAR and MSU-Bench show that Audio-Mind outperforms prior audio-agent baselines, reaching 80.4% accuracy on MMAR and 82.8% accuracy on MSU-Bench. A matched-backbone comparison highlights why this design matters: under strong audio frontends, agentic decomposition can become an orchestration bottleneck when the workflow does not preserve the frontend's holistic audio-grounded judgment. Beyond accuracy, Audio-Mind produces higher-quality, auditable reasoning traces that expose uncertainty, tool evidence, and answer rationales, offering a potential basis for more reliable audio-QA annotation and error analysis.

Real-time synthesis of high-fidelity 3D character motion from audio is a pivotal component for next-generation interactive avatars and virtual assistants. However, most existing approaches are limited to offline processing of complete audio sequences or are constrained to specific domains, rarely handling both speech and music effectively. In this paper, we introduce a novel framework designed to generate continuous, coherent full-body motion from streaming speech and music with low latency. Central to our approach is a unified streaming architecture capable of synthesizing continuous motion from incremental audio inputs. We employ a robust training strategy that enforces strong audio dependency, allowing the model to seamlessly generalize across conversational speech and rhythmic music without requiring explicit domain labels or mode switching. Additionally, we explored Reinforcement Learning to refine the quality of online generation. Furthermore, we bridge reactive animation with intent-driven behavior via a tool-call interface that allows upstream Large Language Models to inject explicit semantic control. By combining this controllability with stream audio-driven synthesis, our framework serves as a plug-and-play solution for transforming voice agents into interactive humanoid avatars. Extensive experiments demonstrate that our method outperforms state-of-the-art realtime baselines in motion quality and synchronization while maintaining the flexibility required for live deployment. Our code, pre-trained models, and videos are available at https://robinwitch.github.io/EchoAvatar-Page.

Audio generation has made significant progress, yet synthesizing unified audio where speech and sounds are naturally composited remains a challenge. Current methods either rely on disjoint pipelines, which fail to capture fine-grained interactions, or require structured inputs and external text rewriting, which limits the flexibility of free-form text prompts. In this paper, we introduce a new task: Free-Form-Text-Prompt-to-Unified-Audio generation, which aims to directly synthesize unified audio containing speech, sound, and their composites from unconstrained natural language. To address this task, we propose PlanAudio, a unified, autoregressive LLM-based framework. First, it simplifies the model architecture by leveraging intrinsic LLM reasoning capability instead of traditional text encoders. Second, it introduces a semantic latent chain-of-thought mechanism, an implicit planning mechanism that bridges high-level semantic understanding and low-level acoustic synthesis. Furthermore, we create PlanAudio-Bench, a specialized benchmark for evaluating composite audio scenarios. We perform evaluations in the scenarios of speech, sound, and their composites. The results demonstrate that PlanAudio generally outperforms the existing pipeline and unified baselines, while staying competitive with models designed for a single scenario. Our analysis further reveals the superiority of semantic latent CoT over other CoT mechanisms and highlights the importance of continuous multi-scenario training curricula.

Speech language models (SpeechLMs) have achieved substantial progress by extending large language models (LLMs) to the speech modality. However, SpeechLM evaluation remains heavily centered on English, limiting reliable assessment of multilingual speech capabilities. Straightforward benchmark transfer through ASR, translation, normalization, and TTS can corrupt language-specific instructions, answer constraints, and spoken forms; for audio understanding, transferring source-language audio also fails to preserve target-language speaker attributes, accents, and paralinguistic properties. To address these limitations, we propose two human-agent benchmark-construction frameworks: one transfers source-language SpokenQA benchmarks into target-language SpokenQA benchmarks, and the other converts target-language ASR corpora into audio understanding benchmarks using transcriptions and speaker metadata. Using these frameworks, we construct and publicly release three Korean speech benchmarks: KVoiceBench and KOpenAudioBench for Korean SpokenQA, and KMMAU for Korean audio understanding, comprising 12,345 samples in total. We evaluate eight recent SpeechLMs and find that English-Korean performance gaps vary substantially across models and task families, and that SpokenQA and audio understanding rankings diverge, revealing complementary weaknesses invisible to English-only evaluation.

While large audio language models (LALMs) have achieved remarkable progress in audio processing at the second- or minute-level scale, understanding hour-level audio remains a fundamental bottleneck. Existing benchmarks predominantly rely on short clips or artificially concatenated segments, failing to faithfully assess LALM capacity for long-range information comprehension in real-world scenarios such as podcasts and lengthy speeches. To address this gap, we introduce VoiceGiraffe, a novel benchmark designed to rigorously evaluate LALMs across diverse real-world scenarios, modalities, and languages under long-context settings. It comprises 1500 curated triplets structured into a dual-level taxonomy of single-hop perception and multi-hop reasoning. We evaluate a broad suite of open-source and proprietary LALMs against human performance. Results underscore three fundamental findings. First, VoiceGiraffe remains highly challenging and far from saturation. Second, we show that no single inference paradigm universally dominates. The E2E inference benefits models with native long-context audio understanding, cascaded caption aggregation stabilizes small models overwhelmed by hour-scale audio, and reasoning-enhanced cascading with external LLM helps weaker models but can bottleneck stronger proprietary systems. Third, we reveal long-range memory persistence as a key bottleneck. LALMs are better at answering questions that require connecting salient causal cues than those requiring sustained tracking of sparse events across long audio, whereas humans show the opposite pattern. These findings position VoiceGiraffe as a challenging and diagnostic testbed for long-form audio understanding, highlighting the need for LALMs with persistent memory and robust long-range aggregation.

Audio tokenizers are fundamental to unifying audio understanding and generation. Understanding requires high-level semantics, while generation demands semantic and acoustic details. Existing unified tokenizers jointly encode both in high-dimensional continuous latents, which increases the modeling burden of Diffusion Transformers (DiTs) for generation. We propose LoSATok, a low-dimensional audio tokenizer for cross-domain audio understanding and generation. Motivated by the observation that 1280-dimensional semantic encoder features are compressible, we introduce a Semantic Bottleneck that compresses them into 128 dimensions, regularized by the proposed time-relation loss for temporal feature consistency. We further design a dual-level semantic supervision method that leverages both high- and low-dimensional semantic signals, enabling the tokenizer to jointly capture semantics and acoustic details within a compact latent space. Experiments on speech, music, and general audio show that SemBo preserves strong low-dimensional semantic capacity and LoSATok retains competitive understanding performance compared with several semantic representations, while consistently improving DiT modeling performance on speech, music, and audio generation. These results demonstrate that LoSATok's low-dimensional representations can effectively support audio understanding and generation. Our code is provided at https://github.com/wxzyd123/LoSATok.

Audio generation has long been fragmented, with speech, music, and sound effects produced by domain-specific models that fail to jointly generate coherent audio scenes from a single description. The key obstacles are insufficient fine-grained supervision for real-world mixed audio and limited acoustic representations for modeling concurrent audio components. We present Dasheng AudioGen, a unified framework for generating general mixed-audio scenes from text. Dasheng AudioGen introduces structured multi-view captions, which explicitly decouple complex acoustic scenes into complementary description views, thereby enabling fine-grained control over audio layers. Furthermore, we employ a high-dimensional unified semantic-acoustic representation as the shared latent space. It injects semantic priors that facilitate cross-modal training convergence, while its high-dimensional feature space provides sufficient capacity to disentangle and fuse concurrent audio components effectively. With these designs, a simple flow-matching DiT achieves high-quality end-to-end audio scene generation. We also establish a comprehensive evaluation pipeline for audio scene generation. Experiments demonstrate that Dasheng AudioGen achieves performance approaching real-world recordings in mixed-audio categories, while remaining competitive with specialized models in single-type generation tasks. Demos are available at https://nieeim.github.io/Dasheng-AudioGen-Web/.

Audio large language models (Audio LLMs) demonstrate strong performance on speech understanding tasks, yet their ability to understand paralinguistic information remains limited. To systematically quantify this issue, we introduce VoxParadox, an adversarial benchmark with 2,000 verified examples, spanning 10 paralinguistic tasks, created with controlled speech synthesis to intentionally mismatch transcript claims and speaking style, enabling direct measurement of speech paralinguistic understanding. Evaluation of a diverse set of Audio LLMs reveals consistently low accuracy on acoustic ground truth and a strong tendency to follow language-implied (incorrect) answers. To understand the cause of this gap, we perform layer-wise probing and find that (i) paralinguistic cues can degrade in deeper encoder layers and at the encoder--LLM interface, and (ii) even when such cues are available in audio tokens, the language model frequently ignores them. To address these problems, we propose Prompt-Conditioned Layer Mixer (PCLM), which adaptively combines information from multiple audio layers based on the input prompt, and pair it with Direct Preference Optimization (DPO) to explicitly prefer acoustically supported options over language-implied alternatives. These methods substantially improve Audio LLM paralinguistic understanding, improving Audio Flamingo 3 from 17.40% to 65.20% on VoxParadox, and from 37.74% to 54.78% on MMSU paralinguistic subset. Our project page is available at https://voxparadox.github.io/.

Audio and omni-modal large language models exhibit impressive cross-modal reasoning capabilities. However, applying standard reinforcement learning post-training algorithms to these models exposes a critical structural vulnerability: methods like GRPO apply uniform policy gradients across all tokens, ignoring their unequal dependence on the non-text source modality. This exacerbates late-stage modality collapse during extended chain-of-thought generation, where models progressively abandon the primary source signal in favor of compressed textual priors, leading to confident but ungrounded hallucinations. To address this, we introduce Modality-Aware Policy Optimization (MAPO), a novel dual-branch reinforcement learning framework. First, MAPO dynamically concentrates the policy gradient on modality-critical tokens using a modality relevance mask, which is derived from the cross-modal differential entropy between an audio-ablated reference and the multimodal policy. Second, it integrates an auxiliary attention loss branch that applies a targeted, temporally scaled penalty to the model's internal attention distributions. This ensures the model actively sustains cross-modal grounding deep into the reasoning trace. Evaluations on complex audio reasoning benchmarks demonstrate that MAPO substantially improves long-horizon reasoning fidelity and multimodal instruction following, achieving highly competitive performance and setting new state-of-the-art results on several key benchmarks among open-weight models. By relying strictly on native statistical signals rather than domain-specific inductive biases, MAPO offers a promising foundation for mitigating epistemic collapse across diverse multimodal systems.

Audio-visual large language models (LLMs) hold strong promise for long-form video understanding, yet their long-video inference is fundamentally limited by the linear growth of video tokens and key-value (KV) caches. We present OmniMem, a memory-efficient streaming framework designed specifically for audio-visual LLMs. Unlike existing compression methods that treat all tokens uniformly, OmniMem introduces a modality-aware memory allocation strategy that separately manages visual and audio contexts, addressing the severe token imbalance between the two modalities. OmniMem further preserves informative and non-redundant KV states through perturbation-aware memory selection, enabling compact memory without sacrificing long-range understanding. To strengthen compression under realistic deployment constraints, we also explore budget-aware fine-tuning, which encourages the model to consolidate useful information into retained memory. Experiments on VideoMME Long, LVBench, and LVOmniBench with video-SALMONN 2+ and Qwen-2.5-Omni show that OmniMem consistently improves over strong training-free compression baselines by 2-4% absolute accuracy under the same memory budgets, with an additional 1-2% gain after fine-tuning.

Current music similarity models typically compute a single, monolithic score, entangling distinct musical dimensions like melody, rhythm, and timbre. This limits user control and interpretability, making it impossible to execute nuanced queries. We introduce MERIT, a framework for learning disentangled, factor-specific music representations tailored to these three core dimensions. To overcome the lack of isolated musical variations in real-world audio, we use a novel training strategy that uses conditional audio generation and source-separated stems to strongly encourage single-factor variation in training data. Our evaluations demonstrate strong factor-wise disentanglement. Each head responds strongly to its intended perceptual dimension while remaining near chance on the others, a representational property that holds across both the synthetic training domain and independent real-world audio.

Recent advances in Large Audio-Language Models (LALMs) have made real-time, streaming spoken interaction increasingly practical. In this setting, reasoning quality and responsiveness are tightly coupled: delaying reasoning until the speech endpoint can improve answer quality but moves deliberation into user-visible response delay, while answering too early risks committing before decisive evidence arrives. We introduce a learnable wait-think-answer control formulation for LALMs. Motivated by the incremental nature of human conversation, the controller decides under partial audio evidence when to wait, when to externalize a compact reasoning update, and when to answer. Using Qwen2.5-Omni-7B as the base model, we construct aligned wait-think-answer traces from spoken reasoning data, train the controller with supervised fine-tuning (SFT), and then apply Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO). The reward combines answer correctness, action validity, update timing, latency synchronization, reasoning quality, and chain consistency, optimizing the complete wait-think-answer trajectory and not the final answer alone. On a six-task synthetic spoken reasoning question answering (SRQA) benchmark, the six-reward DAPO controller improves the row-weighted accuracy from 67.6% to 70.3% while reducing post-endpoint final-think length by 14% under the same Qwen deployment harness. On a 186-item human-recorded Real Audio Bench, a transfer check beyond text-to-speech (TTS)-rendered speech, the controller family remains functional: SFT achieves the strongest accuracy, while the six-reward DAPO controller is the only learned variant whose final-think length falls below the base. These results suggest that a streaming model should learn when to make intermediate reasoning explicit during the audio stream.

Few-shot prompting provides an effective way to adapt auditory large language models to low-resource tasks such as children's speech recognition. However, most auditory large language models are not explicitly trained to perform inference in this demonstration-conditioned format, limiting the extent to which they can benefit from few-shot prompting. To address this limitation, we introduce Few-Shot Aware GRPO (FSA-GRPO), an RL-based post-training recipe that uses a specially designed reward to encourage the model to leverage few-shot demonstrations, thereby strengthening its few-shot adaptation ability. Notably, training with only high-resource adult ASR data improves the model's general few-shot adaptation ability, yielding gains not only in children's speech recognition but also in speech translation and audio understanding. We further study data selection and auxiliary reward weighting to identify an effective training recipe. Our experiments show that when in-domain data are unavailable or cannot be used for training, FSA-GRPO is more effective than direct tuning on related out-of-domain data.

Large audio language models (LALMs) process both speech and environmental acoustic cues, yet struggle to retain non-speech information across multi-turn interactions. The performance gap between semantic (speech) and acoustic (non-speech) understanding remains poorly understood, and the underlying mechanisms of representation and retrieval are still unclear. This work introduces EnvMem, a controlled multi-turn benchmark designed to study this gap and identify the root causes of failures at the representation (i.e., latent embeddings) and retrieval levels (i.e., attention allocation). We further conduct post-hoc interventions to probe representational structure and attention dynamics. Our results reveal representational trajectory drift as the key failure mode, while showing that attention allocation plays a limited role in explaining the observed degradation. Overall, we provide a systematic framework for analyzing and improving non-linguistic memory in long-context LALMs, shedding light on future data and training design for robust acoustic memory modeling.

Large-batch Contrastive Learning (CL), the foundation of modern representation learning, is fundamentally incompatible with the volatile resource constraints of edge devices. This conflict creates a dilemma: small on-device batches degrade model fidelity, while offloading to the cloud incurs unacceptable latency and bandwidth costs. Existing solutions often resort to static model compression, which fails to adapt to the runtime volatility of edge environments. To bridge this gap, we present StreamSplit, a novel framework that makes streaming CL practical across heterogeneous ARM client platforms. StreamSplit resolves the conflict between the continuous nature of ambient audio and the discrete batch requirements of models like CLAP and COLA. We introduce: (1) A distribution-based streaming framework that decouples representation quality from local batch size, using a tractable Hybrid Loss to maintain fidelity despite sparse updates; and (2) An Uncertainty-Guided Adaptive Splitter that uses a lightweight Reinforcement Learning (RL) policy to dynamically partition computation. Uniquely, this policy integrates real-time resource monitoring with embedding ambiguity to optimize the accuracy-latency trade-off on the fly. We evaluate StreamSplit on diverse hardware, from the resource-constrained Raspberry Pi 4 to the high-performance Apple M2. Results demonstrate that StreamSplit reduces per-sample latency by up to 4.7x and cuts bandwidth by 77.1% and energy by 52.3% compared to server-centric baselines. Crucially, it maintains accuracy within 2.2% of server-centric models, proving that adaptive, distributed learning is a viable path for the modern edge ecosystem.

Speech editing and zero-shot Text-to-Speech (TTS) share a similar generative foundation conditioned on speech prompts, yet speech editing demands far stricter local acoustic consistency with surrounding unedited content. While prior work has shown that Supervised Fine-Tuning (SFT) enables TTS models to acquire functional editing capability, this approach remains fundamentally bottlenecked by imperfect paired editing data and coarse-grained optimization signals. To address these limitations, we propose CosyEdit2, a speech editing model built on a two-stage post-training framework that progresses from supervised editing initialization to editing-oriented Group Relative Policy Optimization (GRPO) over target-speech-free data. Extensive experiments demonstrate that CosyEdit2 not only substantially advances speech editing performance, but also unlocks better zero-shot TTS capability, revealing a deeper mutual relationship between the two tasks. Audio samples are available at https://cjy1018.github.io/CosyEdit2.

Audio-language models (ALMs) are increasingly used in real-world applications that require understanding music, from music tutoring and transcription to captioning, recommendation systems, and music production. More broadly, they are becoming an important component of multimodal AI systems that must reason from sensory input rather than text alone. This makes reliable musical perception a critical prerequisite: if a model cannot accurately hear the structure of sound, it cannot be trusted to reason about, teach, transcribe, or act on audio in the real world. Yet existing benchmarks rarely assess one of the most fundamental musical abilities underlying such perception: pitch hearing. Current evaluations tend to probe pitch hearing only indirectly, through higher-level tasks and often in multiple-choice formats, leaving open how reliably ALMs identify fine-grained pitch across instruments, acoustic conditions, and response formats. We introduce PitchBench, an evaluation suite that systematically measures pitch hearing in ALMs. PitchBench comprises 28 experiments spanning absolute and relative pitch perception within sequences and chords, while varying loudness, note duration, sound source, time stretching, background noise, and other acoustic conditions. Tasks range from identifying individual pitches in isolation to tracking a melodic line within a four-part musical texture. Evaluating frontier ALMs, we find that pitch hearing remains highly unreliable: models perform consistently poorly across settings, with accuracy varying sharply by sound source, note duration, and notation format. Current ALMs do not yet possess stable pitch perception, even for controlled synthetic and instrumental stimuli. Alongside the benchmark, we release PitchBench as a Python package containing the evaluation data and data generation tools to support future work on pitch-aware audio-language modeling.

Current open-source diffusion models struggle to generate stable and synchronized audio-visual content, particularly in scenarios demanding complex semantic reasoning. The root cause is that existing methods rely on coarse text embeddings from off-the-shelf encoders to guide audio-video denoising, which discards fine-grained semantics and, critically, lacks a shared long-horizon plan, leading to uncoordinated denoising trajectories and fragile cross-modal alignment. We propose Baton, the first framework that introduces explicit semantic planning into joint video-audio generation. Our key insight is that complementing coarse text guidance with semantically rich, modality-aware planned tokens, jointly reasoned and mutually aligned before denoising, can simultaneously restore fine-grained semantic detail and establish a shared blueprint that coordinates both audio and video denoising trajectories. Concretely, Baton first introduces the VA-Planner, a multimodal language model equipped with dual semantic alignment towers, where learnable queries cross-attend to both video and audio features to produce a pair of semantically aligned video and audio planned tokens as keyframe-level blueprints. These planned tokens are injected into the diffusion backbone via cross-attention layers, providing temporally grounded guidance complementary to coarse text embeddings. Since planned tokens do not share one-to-one spatial-temporal correspondence with diffusion latents, we further propose Relative Semantic RoPE, a relative positional encoding that maps planned tokens and latents into a shared spatial-temporal coordinate frame, enabling each latent to accurately attend to its positionally corresponding semantic cues. Experiments on benchmarks show the effectiveness of Baton both qualitatively and quantitatively.

Audio-language models (ALMs) are increasingly used for audio captioning, question answering, and open-ended audio understanding, but their inference cost remains high when audio inputs are represented as long prefix-token sequences. These audio prefixes consume context budget, increase memory usage, and make deployment harder in resource-constrained or latency-sensitive settings. Existing training-free audio-token reduction methods mainly rely on fixed pooling or score-based pruning. Fixed pooling is content-agnostic, while score-based pruning can preserve isolated salient tokens but discard nearby acoustic context. We propose Local Temporal Bipartite Merging (LTBM), a training-free encoder-space compression method that merges similar nearby audio tokens under an explicit temporal window constraint. Beyond introducing LTBM, we use a controlled Global Merge variant to isolate whether temporal locality itself is a useful inductive bias for audio-token compression. Experiments on AudioCaps, Clotho, and MMAU with Qwen2-Audio show evidence of a task-dependent locality effect: locality-aware merging is more favorable for captioning at several compression settings, especially under stronger compression, while global matching is more competitive for multiple-choice audio understanding. A cross-backbone validation on Audio Flamingo 3 further supports the captioning-side advantage of locality-aware merging under moderate and aggressive compression.

Speech and audio systems operate in inherently non-stationary environments, yet continual learning (CL) research in this domain, especially in the foundation model era, remains fragmented that fail to account for the coupled, geometry-sensitive nature of acoustic representations. Modern speech foundation models operate over highly entangled, continuous representations that jointly encode linguistic, speaker, and paralinguistic factors within a shared latent space. CL is therefore fundamentally about preserving and evolving shared representation structure rather than retaining isolated task knowledge. In this work, we revisit CL for speech from a representation-centered perspective, and introduce a new taxonomy that organizes CL according to how underlying representation geometry evolves under non-stationary acoustic conditions. We further identify key mismatches between current CL assumptions and speech foundation model behavior, and finally outline a set of open challenges and future research directions.

In non-invasive neural language decoding, results can be inflated by sources that are not stimulus-evoked neural evidence: decoder priors, embedding-based metrics, and non-neural structural nuisances such as signal duration. The methodological challenge is therefore attribution: a reported gain is more informative when it can be traced to a specific source. We recast stimulus-locked MEG-to-audio retrieval as an auditing framework that separates apparent performance into three sources - structural shortcuts, window-level stimulus-locked evidence, and cross-window contextual aggregation - and provides a diagnostic for each. Signal-blind Gaussian noise reaches 66.3% Rank@1 (R@1) under variable-length decoding but collapses to near chance once fixed-duration windows and stimulus-identity splits are enforced, isolating structural leakage. Under these controls, fixed-window retrieval recovers measurable MEG-audio discriminability, while an oracle sentence-bucket diagnostic shows that 95.7% of Top-1 errors select the wrong sentence, localising the residual bottleneck to sentence-level competition. We audit this contextual source with Group Context Bias (GCB), an inference-time additive logit bias that pools sentence-consistent evidence across windows while leaving the base retrieval scores and candidate pool fixed. Used as a score-space intervention, GCB makes the contextual source measurable: R@1 shifts from 44% to 52% on Gwilliams and from 22% to 29% on MOUS under the same fixed setting. GCB is auditable under this design: its effect collapses under random-grouping perturbations and vanishes when local evidence is attenuated in MEG or is near chance in EEG, supporting its use as a controlled source-attribution intervention. These results suggest that brain-to-language performance should be source-attributed, not merely reported.

Speech deepfake detection has achieved remarkable success in clean environments but faces significant challenges in complex, real-world scenarios where speech is often mixed with background music or noise. Current state-of-the-art methods rely on semantic features from self-supervised learning (SSL) models, which often fail when processing non-speech or mixed-source audio. In this paper, we first introduce MixFake, a large-scale benchmark dataset designed to simulate diverse acoustic environments with varying SNR levels and mixed authenticity components. To address the "semantic-centric" limitation, we propose a Multi-stream Prompt Tuning framework that injects signal-level priors into SSL backbones. By integrating base, frequency, and texture streams through deep prompt injection, our model effectively captures acoustic artifacts. Experimental results demonstrate that our method significantly outperforms existing baselines, achieving a 0.95% EER in foreground detection and a substantial 7.72% absolute improvement in complex background detection tasks. Our dataset and code are available at https://github.com/saltfish233/MixFake.

Audio deepfakes have improved rapidly recently, yet their effect on human trust in real speech remains unstudied. We present the largest listening study on audio deepfake perception to date, collecting 35,532 judgments from 1,768 participants across 138 text-to-speech and voice conversion systems. Our central finding is a skepticism shift: compared to a 2021 baseline, human accuracy on fake samples barely changed (72.9% to 71.2%), but accuracy on real samples dropped from 72.7% to 64.1%. Participants are not worse at detecting synthesis artifacts; rather, they increasingly distrust authentic speech. Samples generated by commercial and autoregressive language model systems proved hardest to detect (61.3 - 65.9%), while those from traditional seq2seq and flow-matching models remain easier to spot (75.4 - 76.8%). An ML detector that served as a reference point maintained over 94.5% accuracy across all conditions. Our results suggest that the primary threat posed by modern deepfakes may not be mere deception, but the erosion of trust in genuine audio.

The digitization of musical scores plays a crucial role in their preservation and accessibility, yet information retrieval still depends mainly on metadata searches, such as by title or composer. Content based search in music score images remains underexplored compared to text documents, despite its potential value for musicians, musicologists, and educators. This work contributes to the field by first studying which characteristics of a score are most relevant for search and by defining a systematic method to build query datasets from any annotated corpus. We also consider diverse methods for content-based search on music score images, ranging from transcription-based approaches relying on Optical Music Recognition (OMR), to a transcription-free Transformer model trained to recognize queries directly from score images, and a text-prompted Large Language Model. Our experiments evaluate these models on four corpora exhibiting diverse characteristics in terms of dataset size, image quality, and typesetting mechanisms. Overall, each method excels under different conditions: OMR-based pipelines achieve higher in-domain retrieval, whereas transcription-free models handle domain variability more effectively.

Prior attacks on Audio Large Language Models (Audio LLMs) demonstrated that carefully crafted waveform-domain perturbations can force targeted adversarial outputs. As a defense mechanism against these attacks, real-world codec compression preprocessing has been studied to both detect and remove the perturbations. Yet no existing attack has demonstrated robustness against these compressions. We introduce CodecAttack, which optimizes a perturbation in a neural audio codec's continuous latent space rather than directly perturbing the audio waveform. We show that the codec's compression channel, which discards waveform perturbations, transmits perturbations crafted in its own latent space. To further harden the attack across real-world compression channels, we apply multi-bitrate straight-through Expectation-over-Transformation (EoT), all without modifying the target model. Across three realistic Audio LLM deployment scenarios and three target models, CodecAttack achieves an average 85.5% target-substring attack success rate (ASR) on Opus at moderate bitrates, while the waveform baseline trained with identical EoT hardening does not exceed 26% at any bitrate. The attack transfers to held-out codecs, reaching up to 100% ASR on MP3 and 84% on AAC-LC without retraining. A per-band energy analysis shows that the latent perturbation concentrates below 4kHz, exactly where codecs allocate the most bits, while the waveform baseline spreads into higher frequencies that codecs discard. These results demonstrate that lossy compression is not a reliable defense against adversarial audio and that codec-aware attacks pose a practical threat to deployed Audio LLM systems.

Reinforcement learning is a powerful learning paradigm that has spearheaded progress in numerous domains. Its core promise lies in learning through high-level goals without the need for granular labels. However, it still remains elusive in the realm of audio, where it has received substantially less attention than in computer vision or other domains. The key question remains: how can agents learn to listen purely via reward-driven exploration? In this contribution, we present an overview of previous attempts and a new conceptual framework for learning to listen by reward. Our approach depends on the continuous search for novel sound sources. We formulate our framework, discuss open technical challenges, and present a first proof-of-concept implementation that showcases the feasibility of our approach.

Despite rapid advances in automatic speech recognition (ASR) and large audio-language models, robust recognition in real-world environments remains limited by an "acoustic robustness bottleneck": models often lose acoustic grounding and produce omissions or hallucinations under severe, compositional distortions. We propose Mega-ASR, a unified ASR-in-the-wild framework that combines scalable compound-data construction with progressive acoustic-to-semantic optimization. We introduce Voices-in-the-Wild-2M, covering 7 classic acoustic phenomena and 54 physically plausible compound scenarios, and train Mega-ASR with Acoustic-to-Semantic Progressive Supervised Fine-Tuning and Dual-Granularity WER-Gated Policy Optimization. Extensive experiments demonstrate that Mega-ASR achieves significant advantages over prior state-of-the-art systems on adverse-condition ASR benchmarks (45.69% vs. 54.01% on VOiCES R4-B-F, and 21.49% vs. 29.34% on NOIZEUS Sta-0). On complex compositional acoustic scenarios, Mega-ASR further delivers over 30% relative WER reduction against strong open- and closed-source baselines, establishing a scalable paradigm for robust ASR in-the-wild.

Training general-purpose Audio Large Language Models (ALLMs) across diverse datasets is essential for holistic audio understanding, yet it faces significant challenges due to dataset heterogeneity, which often leads to conflicting gradients and slow convergence. Despite its impact, how to explicitly manage this heterogeneity during training remains underexplored, with current practices relying primarily on uniform mixture. In this work, we analyze multi-dataset AudioQA training from a convergence perspective and propose Grouped Sequential Training (GST). GST strategically organizes datasets into affinity-aware groups and introduces them via a progressive scheduling protocol, effectively balancing the stability of parallel training with the efficiency of sequential optimization. To ensure scalability, we develop gradient-based affinity metrics that capture inter-dataset relationships without the prohibitive cost of empirical transferability estimation. Extensive evaluations on 14 AudioQA datasets spanning speech, music, and environmental sounds demonstrate that GST achieves 30--40% faster convergence than standard parallel training while maintaining or even surpassing the performance of mix-all training. Our results provide both theoretical insights and a practical, model-agnostic framework for efficient large-scale ALLM optimization.

The foundational capabilities established by Large Language Models (LLMs) have paved the way for Multimodal Large Language Models (MLLMs), within which Large Audio Language Models (LALMs) are essential for realizing universal auditory intelligence. Despite their remarkable performance, the escalation of LALMs' capabilities has significantly outpaced the development of systemic frameworks to ensure their trustworthiness. This survey provides a comprehensive investigation into the endogenous mechanisms of LALMs, detailing the architectural innovations and alignment algorithms that facilitate emergent reasoning. Specifically, we analyze how the transition to unified end-to-end frameworks and the integration of continuous acoustic signals inherently expand the attack surface. To rigorously evaluate the risks within these paradigms, we establish a comprehensive taxonomy of trustworthiness, categorizing critical vulnerabilities such as cross-modal jailbreaking, latent acoustic backdoors, and biometric privacy leakage. We review the state-of-the-art through six analytical pillars: hallucination, robustness, safety, privacy, fairness, and authentication. The profound imbalance between a mature offensive landscape and underdeveloped defenses further validates the critical trustworthiness gaps and multidimensional risks facing audio-centric intelligence. Finally, we propose a strategic roadmap advocating for "Defense-in-Depth" architectures, causal auditory world modeling, and intrinsic representation engineering to bridge the gap between empirical performance and intrinsically trustworthy audio intelligence. Our project has been uploaded to GitHub https://github.com/Kwwwww74/Awesome-Trustworthy-AudioLLMs.

Modern audio generation predominantly relies on latent-space compression, introducing additional complexity and potential information loss. In this work, we challenge this paradigm with WavFlow, a framework that generates high-fidelity audio directly in raw waveform space without intermediate representations. To overcome the inherent difficulties of modeling high-dimensional and low-energy signals, we reshape audio into 2D token grids through waveform patchify and introduce amplitude lifting to align signal scales, enabling stable optimization via direct x-prediction in flow matching. To capture complex semantic alignment and temporal synchronization, we leverage an automated data pipeline to curate 5 million high-quality video-text-audio triplets, allowing the model to learn fine-grained acoustic patterns from scratch. Experimental results show that WavFlow achieves competitive performance on the video-to-audio benchmark VGGSound (FD_PaSST: 59.98, IS_PANNs: 17.40, DeSync: 0.44) and the text-to-audio benchmark AudioCaps (FD_PANNs: 10.63, IS_PANNs: 12.62), matching or exceeding the performance of established latent-based methods. Our work demonstrates that intermediate compression is not a prerequisite for high-quality synthesis, offering a simpler and more scalable alternative for multimodal audio generation.

Recent diffusion-based methods have achieved impressive progress in video content manipulation. However, they typically ignore the accompanying audio, leaving the audio disjointed from the edited results. In this paper, we propose InstructAV2AV, the first end-to-end framework for instruction-guided audio-video joint editing. We first develop a scalable data synthesis pipeline and construct InsAVE-80K, the first large-scale audio-video editing dataset with high-quality source-to-target pairs. With this data foundation, we adapt an audio-video generation backbone to leverage its robust priors. We concatenate the audio-video input with noisy latent codes to anchor the source context, propose the source-instruction gated attention to improve instruction following and content preservation, and introduce a two-stage training strategy to effectively transfer these pre-trained priors. Extensive experiments demonstrate that InstructAV2AV outperforms state-of-the-art methods across 11 metrics spanning three aspects on two evaluation sets, highlighting its potential for controllable content creation. Project page: https://hjzheng.net/projects/InstructAV2AV/.

The integration of audio modality into Large Audio Language Models (LALMs) significantly expands their attack surface. Existing jailbreak paradigms predominantly treat audio as a carrier for malicious payloads, relying on semantic optimization, acoustic parameter control, or additive perturbation to embed harmful content into the audio signal. In this work, we challenge this necessity and propose a new paradigm in which the role of audio shifts from content injection to safety alignment interference. We reveal that LALM safety alignment can be compromised solely by specific Acoustic Latent Semantics (ALS), the underlying paralinguistic features intrinsic to the priors of audio generative models. Distinct from previous works that leverage explicit acoustic parameters to merely style malicious audio, we demonstrate that interference audio, benign in content but infused with specific ALS, can serve as a universal jailbreak trigger. Leveraging this insight, we propose the Acoustic Interference Attack (AIA), which decouples the attack payload from the audio. Specifically, AIA employs a set of universal, instruction-neutral interference audio, enabling standard malicious text queries to bypass safety alignment without instance-specific optimization. Extensive experiments on 10 LALMs across five datasets demonstrate that AIA achieves the state-of-the-art attack success rate. Furthermore, our interpretability analysis uncovers the inference path drift induced by AIA and identifies the inherent effective patterns within ALS, revealing the fundamental vulnerability of cross-modal alignment in LALMs.

Stable Audio 3 is a family of fast latent diffusion models (small, medium, large) for variable-length audio generation and editing. Since our models can generate several minutes of audio, variable-length generations are key to avoid the cost of producing full-length generations for short sounds. We also support inpainting, enabling targeted audio editing and the continuation of short recordings. Our latent diffusion models operate on top of a novel semantic-acoustic autoencoder that projects audio into a compact latent space, enabling efficient diffusion-based generation while preserving audio fidelity and encouraging semantic structure in the latent. Finally, we run adversarial post-training to both accelerate inference and improve generation quality, reducing the number of inference steps while improving fidelity and prompt adherence. Stable Audio 3 models are trained on licensed and Creative Commons data to generate music and sounds in less than a 2s on an H200 GPU and less than a few seconds on a MacBook Pro M4. We release the weights of small and medium, that can run on consumer-grade hardware, together with their training and inference pipeline.

We investigate Counterfactual Video Foley Generation, which aims to adopt a sound-source identity that contradicts the visual evidence while remaining temporally synchronized to a silent video. Existing Video&Text-to-Audio (VT2A) models struggle with this, often remaining anchored to the visually implied sound source when video and text contents disagree. We present ConterFlow, an inference-time dual-phase sampling scheme for pretrained flow-matching VT2A models. Phase 1 builds a video-derived temporal structure while suppressing the visually implied source; Phase 2 drops video conditioning to focus entirely on shaping audio timbre toward the target prompt. ConterFlow substantially improves counterfactual Video Foley generation compared to naive negative prompting and state-of-the-art baselines. To evaluate replacement quality, we propose a metric leveraging a text-audio co-embedding space to measure both target-prompt evidence and residual visually implied source leakage. Video demonstrations and code are available at https://gyubin-lee.github.io/counterflow-demo/

Recent Singing Voice Synthesis (SVS) advances enable highly realistic but potentially malicious AI covers, making singing voice deepfake detection (SVDD) crucial. Self-Supervised Learning (SSL)-based detectors achieve state-of-the-art performance by fine-tuning speech SSL backbones to capture singing-specific spoof artifacts. Existing adversarial attacks often fail against SSL-SVDD, creating a false impression of inherent robustness. We reveal this stems from two challenges. First, at the objective level, attacks optimize cross-entropy on local surrogates, crossing surrogate-specific boundaries rather than suppressing shared spoof evidence. Second, at the method level, attacks follow the surrogate's dominant gradient direction. In SSL-SVDD, this aligns with fine-tuned artifact-sensitive directions, limiting transferability to unseen detectors - a geometric failure we term the Linearity Trap. To properly evaluate robustness, we propose MARS (Meta-Adversarial Regression of Semantics), a transfer-based black-box framework tailored to SSL-SVDD. Structurally, MARS shifts to hypothesis-evidence manipulation by constructing a natural semantic anchor from the pre-trained SSL space and an artifact anchor from the fine-tuned space. Algorithmically, MARS escapes the Linearity Trap via bi-level optimization: the inner stage induces tangential exploration, while the outer stage guides the audio toward the natural semantic manifold. Experiments on the CtrSVDD benchmark show MARS improves Attack Success Rate (ASR) in in-distribution transfer (13%), out-of-distribution transfer (10%), and cross-task evaluation (36%), highlighting the urgent need for robust SVDD systems.

There are not enough established benchmarks for the task fo speech summarization. Creating new benchmarks demands human annotation, as LLMs could embed systemic errors and bias into datasets. We test ten annotation workflows varying input modality (audio, transcript, or both) and the inclusion of editing (self or peer-editing) to investigate potential quality tradeoffs from using human annotators to summarize audio. We compare human audio-based summaries to human transcript-based summaries to track the impact of the different information modalities on summary quality. We also compare the human outputs against four LLM benchmarks (three text, one audio) to examine whether human-written summaries are less informative than highly fluent automated outputs. We find that audio-based summaries are less informative and more compressed than transcript summaries. However, iterative peer-editing with audio mitigates this difference, enabling audio-based summaries to be as informative as their transcript counterparts and LLM summaries. These findings validate iterative peer-editing among human annotators for the creation of benchmarks informed by both lexical and prosodic information. This enables crucial dataset collection even in setting where transcripts are unavailable.

Weakly labeled datasets such as AudioSet have driven recent progress in audio tagging. However, annotation quality varies across sound classes. Labels may be incomplete, ambiguous, or unreliable, which introduces class-dependent supervision bias during optimisation. The issue becomes harder as real and generated audio are increasingly mixed in training, and generated samples do not always match their intended semantic labels. Prior work mainly addressed unreliable supervision from missing-positive labels, while this paper targets three other sources of unreliable supervision: spurious additions, misassignments between similar classes, and weakened label evidence. These effects introduce class-dependent optimisation bias that is not explicitly modeled by most existing methods. To bridge this gap, the paper proposes a Class-wise Supervision Unreliability (CSU) framework that controls supervision strength at the class level during training. CSU learns a separate unreliability parameter for each class and down-weights less reliable supervision without changing the model architecture or inference process. To support evaluations, this paper also introduces ESC-FreeGen50, a manually verified benchmark of 50 sound classes that combines real and generated audio. Experiments on controlled benchmarks and AudioSet show that CSU improves robustness across different architectures and different sources of supervision unreliability. The results indicate that explicit class-wise modeling of supervision unreliability is an effective and practical strategy for robust audio tagging under large-scale weakly labeled training. Code and data are available at: https://github.com/Yuanbo2020/CSU

The landscape of joint audio and video generation has been fundamentally transformed by the advent of powerful foundation models. Despite these strides, achieving cohesive multimodal customization for the simultaneous preservation of visual identities and vocal timbres across multiple interacting subjects remains largely underexplored. To bridge this gap, we present Omni-Customizer, an end-to-end framework targeted at the precise binding and seamless fusion of multimodal identity information. Specifically, we introduce an Omni-Context Fusion (OCF) module that effectively enriches the base textual prompt with dense, multimodal identity cues, along with a Masked TTS Cross-Attention (MTP-CA) mechanism explicitly designed to prevent the severe "speech leakage" problem. Within this architecture, we propose Semantic-Anchored Multimodal RoPE (SA-MRoPE) to anchor visual and audio reference tokens, along with TTS embeddings, to their corresponding semantic descriptions, enabling structured multimodal fusion and robust identity binding. Furthermore, we devise a comprehensive training strategy that incorporates interleaved audio-video scheduling to rapidly adapt the audio branch to multilingual scenarios without degrading foundational priors, and a progressive in-pair to cross-pair curriculum to facilitate the learning of high-level and robust identity features. Extensive experiments demonstrate that Omni-Customizer achieves state-of-the-art performance in dual-modal customized generation, excelling across visual identity similarity, timbre consistency, precise audio-video synchronization, and overall video-audio fidelity.

Latent diffusion models have emerged as the dominant paradigm for many generation tasks including audio generation such as text-to-audio, text-to-music and text-to-speech. A key component of latent diffusion is an autoencoder (VAE) that compresses high-dimensional signals into a low frame rate continuous representation that is conducive for downstream prediction. Regularizing these VAEs is challenging, as there is a trade-off between over-regularized (poor output quality) and under-regularized (difficult to predict) latent representations. We propose a framework for studying this trade-off through compression and train Audio VAEs at specific bitrates via target-KL regularization. This allows direct comparison to well-studied discrete neural audio codec models, and the construction of rate-distortion curves for audio VAEs. We evaluate the impact of target-KL regularization on text-to-sound generation and find that sweeping compression rates is helpful in identifying the optimal generation setting.

Motion-centric video editing remains difficult for large generative video models, which often respond well to appearance changes but struggle to produce specific, localized actions or state transitions in an existing clip. We introduce Sound Sparks Motion, a training-free framework that enables motion editing in an audio-visual video generation model by tuning its internal multimodal conditioning signals at test time. Rather than modifying model weights, our method tunes only two lightweight variables: an audio latent derived from the source video and a residual perturbation in the text-conditioning. We find that this combination can encourage motion edits that the underlying model often struggles to realize under prompt-only control. Since there is no direct way to evaluate temporal alignment between text and motion, we guide the tuning process using a vision-language model that provides feedback indicating whether the intended motion appears in the generated video. This simple supervision yields an effective semantic objective for motion editing, while regularization and perceptual-temporal constraints help preserve content and visual quality. Beyond per-video tuning, we show that the learned latent controls are transferable across videos, suggesting that they capture reusable motion-edit directions rather than overfitting to a single example. Our results highlight multimodal conditioning tuning, particularly through the audio pathway, as a promising direction for motion-aware video editing, and suggest that test-time tuning can serve as a lightweight probing mechanism that helps reveal latent motion controls embedded in the model's multimodal conditioning. Code and data are available via our project page: https://amirhossein-razlighi.github.io/Sound_Sparks_Motion/

Voice agents increasingly require reliable tool use from speech, whereas prominent tool-calling benchmarks remain text-based. We study whether verified text benchmarks can be converted into controlled audio-based tool calling evaluations without re-annotating the tool schema and gold labels. Our dataset-agnostic framework uses text-to-speech, speaker variation, and environmental noise to create paired text-audio instances while preserving the original dataset annotations. Based on extensive evaluation of 7 omni-modal models on audio-converted versions of Confetti and When2Call, our framework demonstrates that the performance is strongly model- and task-dependent: Gemini-3.1-Flash-Live obtains the highest Confetti score (70.4), whereas GPT-Realtime-1.5 performs best on When2Call (71.9). On Confetti, the text-to-voice gap ranges from 1.8 points for Qwen3-Omni to 4.8 points for GPT-Realtime-1.5. A targeted analysis of failure cases demonstrates that degradations most often reflect misunderstandings of argument values in the speech. Considering real-world deployment scenarios, we further report text-only results, an ambiguity-based reformulation stress test, and a reference-free LLM-as-judge protocol validated against human preferences. Notably, we find that open-source Qwen3 judges with at least 8B parameters exceed 80% agreement with proprietary judges, supporting privacy-preserving evaluation. Overall, our framework provides a verifiable and reproducible first-stage diagnostic that complements purpose-built audio corpora.

As audio-first agents become increasingly common in physical AI, conversational robots, and screenless wearables, audio large language models (audio-LLMs) must integrate speaker-specific understanding to support user authorization, personalization, and context-aware interaction. This requires modeling who is speaking, how the voice sounds, and how recording conditions affect speaker cues. Conventional speaker verification systems provide strong scalar scores but little linguistic evidence, while current audio-LLMs and speaker-aware language models have limited ability to organize speaker information beyond binary labels or descriptive profiles. We present SpeakerLLM, a speaker-specialized audio-LLM framework that unifies single-utterance speaker profiling, recording-condition understanding, utterance-pair speaker comparison, and evidence-organized verification reasoning within a natural-language interface. We construct verification-reasoning targets and a decision-composition policy that separate profile-level evidence from the final same-or-different decision and organize recording condition, profile evidence, and the decision into a structured trace. At its core, SpeakerLLM uses a hierarchical speaker tokenizer designed to capture multiple granularities of speaker evidence. Utterance-level speaker embeddings summarize identity and profile-level cues, whereas frame-level speaker features preserve fine-grained acoustic descriptors. Experiments show that SpeakerLLM-Base improves speaker-profile and recording-condition understanding over general audio-LLMs, while SpeakerLLM-VR preserves strong generated-verdict accuracy and produces decision traces grounded in the supervised verification reasoning schema. We will release the metadata-enriched supervision dataset and target-construction code for reproducibility.

While existing text-to-speech (TTS) models exhibit high expressiveness, fine-grained control over composite instructions remains challenging due to the structural mismatch between discrete textual intents and continuous acoustic realizations. Inspired by human cognitive decoupling, we introduce AgentSteerTTS, a multi-agent closed-loop framework designed for intent-faithful expressive control of composite instructions. First, in our framework, an adversarial disentanglement agent mitigates speaker-emotion leakage by learning separable identity and emotion-prosody subspaces with leakage-suppressing regularization. Next, a Dual-Stream Anchoring Controller grounds abstract intents using a large-scale acoustic prototype library: a Retrieval Agent selects expressive anchors, while a Synthesis Agent fuses them into continuous control vectors via gated attention. Finally, a Fast-Slow Feedback Agent refines output intensity through latent gradient correction and resolves semantic-acoustic mismatches using high-level perceptual critique. Experiments on a composite-instruction benchmark and public test sets show that AgentSteerTTS yields consistent and significant improvements to the baselines, demonstrating the effectiveness of the proposed method.

Audio self-supervised learning (SSL) aims to learn general-purpose representations from large-scale unlabeled audio data. While recent advances have been driven mainly by generative reconstruction objectives, contrastive approaches remain less explored, partly due to the difficulty of designing effective audio augmentations and the large batch sizes required for contrastive pre-training. We introduce \textbf{AudioMosaic}, a contrastive learning-based audio encoder for general audio understanding. During pre-training, AudioMosaic constructs positive pairs by applying structured time-frequency masking to spectrogram patches, which reduces memory usage and enables efficient large-batch training. Compared with generative approaches, the AudioMosaic encoder learns more discriminative utterance-level representations that demonstrate strong transferability across datasets, domains, and acoustic conditions. Extensive experiments show that AudioMosaic achieves state-of-the-art performance on several standard audio benchmarks under both linear probing and fine-tuning. We further show that integrating the pretrained AudioMosaic encoder into audio-language models improves performance on audio-language tasks. The code is publicly available in our \href{https://github.com/HanxunH/AudioMosaic}{GitHub repository}.

Bioacoustic recognition requires fine-grained acoustic understanding to distinguish similar-sounding species. However, many large-scale data repositories such as iNaturalist are weakly annotated, often with only a single positive species label per recording, making supervised learning particularly challenging. Inspired by advances in computer vision, recent approaches have shifted toward self-supervised learning to capture the underlying structure of audio without relying on exhaustive annotations. In particular, masked autoencoders (MAE) have shown strong transferability on massive audio corpora, yet their effectiveness in more modest bioacoustic settings remains underexplored. In this work, we conduct a systematic study of MAE pretraining for species classification on iNatSounds, analyzing the impacts of pretraining data scale, domain specificity, data curation, and transfer strategies. Consistent with prior work, we find that models pretrained on diverse general audio data achieve the best transfer performance on iNatSounds. Contrary to observations from large-scale audio benchmarks, we find that (1) additional masked reconstruction pretraining on domain-specific data provides limited benefits and may even degrade performance relative to off-the-shelf models, and (2) selective data filtering offers a negligible advantage when the overall data scale is limited. Our results indicate that, in moderate-sized fine-grained bioacoustic settings, pretraining scale dominates objective design. These findings further clarify when MAE-based pretraining is effective and provide practical guidance for model selection under limited supervision.

Audio large language models (Audio LLMs) exhibit systematic failures in transcribing code-switching speech despite strong multilingual capabilities. Focusing on English-Mandarin, we identify three failure modes: language omission, translation-instead-of-transcription, and hallucination. We apply Direct Preference Optimization (DPO) to align models, constructing preference pairs in which chosen responses preserve mixed-language content while rejected responses mimic failure patterns. Training three Audio LLMs on 100K pairs (570 hours), we observe consistent behavioral shifts: models learn to preserve language composition rather than translating when prompted for transcription. This alignment yields MER reductions up to 89.6% (in-distribution) and 20.0% (out-of-distribution). Our findings suggest DPO can effectively elicit correct code-switching transcription behavior from multilingual Audio LLMs.