Audio Language Models

In-vehicle music can serve as an adaptive interface to enhance driver experience, attention, and well-being. We present Drive-to-Music, a context-aware system that generates music in real time from multimodal driving signals. Using dashcam imagery and vehicle telemetry, the system extracts scene semantics and driving context, maps them to high-level musical descriptors, and conditions generative audio models to produce contextually aligned soundtracks. The architecture combines perception and generative components to translate visual and kinematic inputs into structured musical attributes and synthesize audio with low latency. It supports smooth transitions as driving conditions evolve, and to ensure robustness and deployment readiness, we incorporate constraint-based controls and safety checks across the generation pipeline. Our results demonstrate the feasibility of real-time, context-aware music generation in automotive settings, providing a foundation for personalized and adaptive in-vehicle audio experiences.

Visual content is the dominant medium of communication, yet without audio descriptions (ADs), it remains inaccessible to blind and low-vision people. ADs narrate context-relevant visual events during natural audio pauses. Manually creating ADs is expensive, limiting coverage to a small fraction of available content. Most existing automatic AD generation methods frame the task as video clip captioning, requiring ground-truth timestamps and additional context cues such as character databases. Current benchmarks reinforce this framing, consisting of short video segments paired with automatic or task-mismatched annotations. We introduce StrAD, a benchmark for long-form AD generation on full-length videos spanning diverse genres such as movies, documentaries, short films, performances, and video games. We reformulate AD generation as streaming dense video captioning. Our approach processes full-length videos with a sliding window, inserting ADs into existing transcripts without ground-truth timestamps, and supports both fine-tuned models and zero-shot prompting of vision-language models. On the segment-level task with given timestamps, our fine-tuned StrAD-FT sets the state of the art on CMD-AD with 36.3 CIDEr (+10.0 over Shot-by-shot), establishes a reference point on StrAD (51.0 CIDEr), and remains competitive on MAD-Eval at 24.9 CIDEr. On the full-video streaming task, StrAD-FT reaches a SODA score of 2.4 against 1.1 for our zero-shot baseline StrAD-Zero, though both exhibit limitations in temporal localization and narrative coherence. While prior work has tackled full-video AD generation in an offline, multi-stage fashion, ours is the first streaming approach, generating ADs on the fly without ground-truth timestamps. StrAD makes progress on full-video AD generation measurable, a prerequisite for scaling accessibility.

Generating coherent audio scenes that simultaneously blend speech, music, and sound effects remains a significant challenge. Current approaches typically rely on a disjointed pipeline where a frozen, decoupled text encoder feeds a separate audio decoder, limiting cross-modal optimization and leading to poor speech intelligibility. To overcome these limitations, we introduce MiDashengLM-Gen, an end-to-end framework that couples a pre-trained Large Language Model (LLM) with per-token conditional flow matching for autoregressive, variable-length mixed-audio scene generation. MiDashengLM-Gen represents a first approach for general text-to-audio generation with one end-to-end trained model. Empirical evaluations demonstrate that MiDashengLM-Gen drastically improves speech intelligibility over existing unified models. On the Seed-TTS benchmark, English Word Error Rate (WER) drops from 12.15% to 2.79%, approaching the performance of dedicated Text-to-Speech (TTS) systems (1.24%). Furthermore, the framework extends effectively to multilingual settings, yielding highly competitive multilingual WERs compared to existing baselines. Lastly, the model maintains competitive mixed-audio generation quality on the MECAT benchmark. Code and checkpoints are available at https://github.com/xiaomi-research/midashenglm-gen and https://huggingface.co/mispeech/midashenglm-gen, and the demo page is available at https://xingws.github.io/midashenglm-gen-demo/.

Recent advances in model design and self-supervised audio representations have improved speech and audio understanding, yet infant-centered naturalistic recordings remain challenging due to limited labeled data, low signal-to-noise ratio, and cross-family domain shifts. We present a family-conditioned, multi-tier audio tagger that combines a LoRA-finetuned Whisper encoder with a lightweight, target-speaker-aware Transformer for long-context inference and framewise prediction across tiers. To improve temporal coherence, we incorporate a simple sequence-level smoothing loss, and to enhance robustness across households, we introduce a factorized speaker-token design with a shared tier token and a learned family-specific offset, reducing family bias and promoting generalizable representations. Together, these choices enable efficient and effective infant-centered audio tagging of daylong audio recordings in home environments.

A common approach to adding audio to a vision-language model is to train or adapt a large omni-modal system. We show that a lightweight alternative can be highly effective for music audio-visual question answering (AVQA). Qwen-MusicAVQA-7B connects a frozen Whisper encoder to Qwen2-VL-7B-Instruct through learned linear projections. The same frozen encoder processes both the video's music track and a TTS-spoken question through separate projectors, while the language model fuses visual frames, music, and question audio through pretrained self-attention, with no task-specific fusion network. On MUSIC-AVQA, our system reaches 96.0% +/- 3.9% accuracy across three independent training seeds on the 7,402-question available-video test subset. Our central finding is that downstream accuracy tracks how much fine-grained local temporal information the audio representation preserves. In a matched 32-token comparison, a stride-pooled Whisper frame sequence outperforms a globally pooled PANNs representation expanded to the same budget by 26 percentage points, even though PANNs sees at least as much audio and uses a far larger projector. The effect is not simply sequence versus vector: within Whisper alone, reducing temporal resolution at a fixed token budget costs a comparable amount. Under matched data and inputs, fine-tuned Qwen2.5-Omni-7B reaches 80.9%, against 95.9% for our 30 s variant; because the systems differ in backbone and adaptation, this is a system-level comparison. Accuracy remains high on sampled head and tail splits of the rephrased MUSIC-AVQA-R benchmark (96.5% and 95.6%). Because both encoders stay frozen and the music features are cached, the entire adaptation is cheap to train: the complete two-stage AVQA run takes approximately 5 hours on a single A100 80GB, and every run reported here fits on that one GPU.

We present DINO-A, an adaptation of self-distillation from vision to general audio representation learning. While DINO has become a canonical method in self-supervised vision and prior audio work has explored latent prediction (BYOL-A) and masked modeling (Audio-MAE, BEATs), no prior work has brought canonical DINO to general audio classification in the way BYOL-A brought BYOL. DINO-A retains DINO's multi-crop, EMA teacher, and high-dimensional projection, replacing only the input modality and augmentations with log-mel spectrograms and the BYOL-A v2 augmentation block. We pretrain three backbones, two Vision Transformers with 8x8 and 16x16 patches and a convolutional encoder, on FSD50K and evaluate them with linear probing on ESC-50, Speech Commands v2, UrbanSound8K, and GTZAN. Three findings characterize the resulting representations. Patch resolution within the Vision Transformer family has consistent effect on representation quality, with smaller patches winning across all four tasks. The choice between Vision Transformer and convolutional backbone interacts with task type: convolutional networks lead on speech while Vision Transformers lead on environmental sounds and music. Under identical pretraining and evaluation conditions, DINO-A and BYOL-A v2 differ by 11.96 percentage points on average, and we trace this difference to two mechanisms: the interaction between DINO's high-dimensional projection space and FSD50K's limited scale, and the additional cost of multi-crop augmentation, which DINO uses but BYOL-A v2 does not. The high-dimensional projection space, central to DINO's success in vision, becomes a liability at FSD50K scale.

Audio effects (Fx) representation learning plays a key role in intelligent music production, including automatic mixing and Fx style transfer. Existing methods typically rely on dry or nearly dry references for effect modeling, yet truly unprocessed audio is rarely available in practice, as real recordings inevitably reflect the microphone, room acoustics, and preceding signal processing. Instead of pursuing absolute effect encodings, we argue that the relative effect distance between audio signals is more meaningful for real-world music production. Motivated by this, we propose RelFx, a contrastive learning framework that learns relative effect transformations from general audio collections without requiring dry references during representation training. Our approach uses a dual-branch Siamese encoder equipped with cross-attention and differential gating fusion to infer the shared effect transformation from a reference clip and an effect-processed, content-related clip. We further propose an antisymmetric fusion variant for bidirectional effect encoding, such that swapping the input order directly produces a nearly sign-reversed embedding, a property not explored in earlier work. Moreover, our dry-reference-free formulation eliminates the reliance on dry multitrack datasets and enables training on effect-bearing audio. Experiments on Fx style transfer demonstrate state-of-the-art performance under the standard Fx-Encoder++ MUSDB18 evaluation protocol, consistently outperforming existing approaches across all four instrument categories.

Automated Text-to-Speech (TTS) evaluation methods (Mean Opinion Score (MOS) predictors and Audio Large Language Models (Audio-LLM) judges) are expected to reflect human perception, yet it is unclear how well they capture the distinct aspects of speech that listeners actually perceive. We deconstruct "naturalness" into a linguistically grounded annotation schema spanning 10 distinct perceptual dimensions, and use it to construct the first dimension-level meta-evaluation benchmark for TTS, comprising 860 utterances annotated by trained linguist raters. Results from benchmarking four MOS predictors and four Audio-LLM judges reveal that MOS predictors collapse onto acoustic signal quality, while Audio-LLM judges show selective, prompt-dependent detection that does not generalise across all dimensions. Neither class reliably captures a breadth of linguistically structured speech errors. Our dataset, annotation schema, and evaluation code are publicly released to support more targeted and interpretable TTS evaluation.

Bioacoustic recordings are often degraded by ambient noise, which complicates the analysis of weak or noise-overlapped vocalizations. Convolutional neural networks, particularly U-Net architectures, have shown a strong denoising performance in speech and music processing. However, their direct application to bioacoustic signals is limited by the scarcity of clean training data. To address this issue, we propose a training set synthesis approach and develop a supervised denoising model that predicts a complex ratio mask in the time-frequency domain. The model leverages ridges, or frequency contours, that represent the fundamental frequency together with one or more harmonic partial components of vocalizations. These ridges are used both for the synthesis of training sets and to design a loss function that assigns higher weights to the ridge regions (ridge-guided loss function). This weighting step helps the network better preserve vocalization details during denoising. As a case study, we evaluate our approach using ultrasonic vocalizations (USVs) recordings of house mice, which are widely studied in behavioral biology and neuroscience. In actual field recordings, the proposed method enhances fundamental and harmonic partial ridge tracking compared to our previous signal-processing approach. In addition, a classifier trained on denoised data improves USV classification on out-of-sample, noisy recordings from wild and domesticated mice compared to classifiers trained on noisy recordings. Our proposed method also substantially improves the scale-invariant signal-to-distortion ratio on synthetic testing data across a wide range of input signal-to-noise ratios. Although we focus on USVs, the proposed approach should be broadly applicable to other bioacoustic signals with trackable ridges, and thus enables ridgebased training set synthesis and denoising.

Narrating the visual component of movies is known as audio description. It is a narrative technique designed to enable blind and visually impaired individuals to follow the story. However, it is far more constrained than most narratives: the descriptions not only need to convey the story in the movie, but they must also fit into gaps between dialogue and they need to conform to guidelines that exist in each region. In this work, we compare audio description created in the United Kingdom against audio description created in the United States. We use guidelines written for these two regions, alongside the impressions from a practitioner in the field, to motivate specific hypotheses about the differences. We test these hypotheses against our pre-existing corpus, which provides both human-authored American and British audio description for each of 206 movies. Results provide quantitative evidence to uphold all tested hypotheses, including differences in lexicon, the use of the progressive aspect, the use of passive constructions, the use of subjective adjectives and modifiers, when characters are named, how scenes are cued, and degree of overlap with movie dialogue and music. Our work offers a quantitative lens into the narrative technique of audio description.

Audio Description (AD) is a verbal narration of key visual content in videos, enabling access for visually impaired audiences. Unlike standard video captioning, AD is a structured editorial task: descriptions must be inserted into gaps in dialogue and must convey only what is needed to understand the narrative being told. However, existing approaches formulate AD generation in an artificial setting where both the content and timing of descriptions are pre-specified, reducing the task to clip-level captioning. They further rely on noisy transcription and alignment pipelines, and lack the rich parallel data required for modeling narrative context. We introduce a new formulation of AD generation in which models must jointly decide what to describe and when to do it. To support this, we present REFRAMED, a high-quality dataset of 2,023 videos that span 3,302 scenes from 206 movies, with professional AD transcripts (both American and British versions), professional subtitles and aligned screenplays. We also provide a manually curated challenge set that pairs full movies with multiple AD references, together with evaluation protocols that leverage dialogue gaps and multi-reference comparisons. Experiments with state-of-the-art AD systems and multimodal LLMs show that they outperform trivial baselines but fall far short of expert human performance. Our dataset and benchmark establish a new foundation for research on video understanding.

Recent advances in speech synthesis and audio generation have made high-fidelity acoustic forgery low-cost and difficult to attribute, enabling a realistic attack scenario in which speech and background audio are independently manipulated over otherwise authentic video. Yet existing research either focuses on visual manipulation, addresses speech detection in isolation, or conflates speech and non-speech audio as a single undifferentiated audio stream, overlooking the distinct forensic challenges posed by background audio. This conflation is consequential: the two acoustic components arise from fundamentally different generative mechanisms, exhibit distinct artifact profiles, and pose different challenges to detection systems. We introduce MADBench, the first benchmark that treats speech and environmental audio as distinct acoustic components, enabling component-aware evaluation of audio deepfake detection across independently manipulated forgery sources. We benchmark representative state-of-the-art detectors and multimodal large language models under a unified protocol. Our experiments reveal that environmental audio manipulation is more detectable than synthetic speech across general-purpose encoders, while existing pretrained detectors fail on both acoustic components, and manipulated environmental audio asymmetrically degrades speech deepfake detection, findings entirely invisible under the single-label paradigm of prior benchmarks. MADBench establishes a rigorous foundation for future research into robust, component-aware audio deepfake detection.

Text-conditioned general audio generation is moving beyond isolated speech, music, and sound-effect synthesis toward a single model that can compose them into controllable, coherent audio scenes. This unified setting is particularly challenging: heterogeneous components impose conflicting structural requirements on a shared backbone, while a complex mixed scene may contain locally distinct or overlapping content that demands fine-grained adaptation within the same clip. Existing audio mixture-of-experts (MoEs) mainly route at the domain level, while token-wise routing overlooks the local continuity inherent to acoustic signals. We propose SonicWeave, a flow-matching model for unified audio scene generation. At its core is a chunk-routed MoE with a conflict-gated prior-evidence routing mechanism (CPE-MoE). CPE-MoE routes contiguous acoustic chunks by combining a global prior that encodes the structured text condition and diffusion phase with local evidence from the evolving acoustic state. A learned conflict gate favors the prior when local states are unreliable, while allowing local evidence to influence routing when a region departs from the global scene context. SonicWeave supports speech, music, sound effects, singing, and their fine-grained mixtures with a single set of weights. Across TTS, TTA, and TTM benchmarks, SonicWeave consistently improves over controlled Dense and Base-MoE baselines. Complex-scene evaluation further demonstrates improved compositional quality, while routing analyses reveal content-dependent expert specialization across diffusion phases. These results suggest that temporally coherent, prior-evidence routing is an effective conditional-computation strategy for unified audio generation. Project page: https://caiyunrui.github.io/SonicWeave.

Understanding dynamic sound sources requires jointly determining what produces a sound, where the source is located, and how it moves over time. Yet existing audio-language models often represent clips as global acoustic events, while vision-language models lack the spatial audio cues needed to localize and track individual sources. To evaluate this missing capability, we introduce ST-OmniQA, a spatio-temporal audio-visual question-answering benchmark built from panoramic videos paired with synchronized first-order Ambisonics (FOA) audio of moving sound sources. It contains 40K videos and 400K question-answer pairs organized into four capability levels covering sound-event recognition, direction of arrival, source distance, motion trajectories, and temporally grounded audio-visual reasoning. Building on this benchmark, we propose ST-Omni-R1, which integrates FOA-derived semantic and trajectory representations with panoramic visual context and is trained through progressive curriculum learning and reasoning-tree reinforcement learning. ST-Omni-R1 achieves 77.83% average semantic accuracy across the four levels, compared with 37.28% for the best evaluated baseline. Results on three public spatial-audio benchmarks further indicate that its learned spatial and motion representations transfer beyond ST-OmniQA.

Brain-to-audio reconstruction is limited by \emph{prior domination}: when a pretrained generator is conditioned on a weak neural signal, it produces realistic but stimulus-inaccurate audio. We introduce RAG-Audio, which decodes fMRI into a semantic audio embedding, retrieves a matching real-audio exemplar, and initializes the frozen generator's sampling trajectory from that exemplar while retaining the decoded embedding as conditioning. On Brain2Music, RAG-Audio improves 10-way stimulus identification from $0.14$--$0.18$ for direct generation, near the $0.10$ chance level, to $0.40$--$0.43$, comparable to retrieval. It also reduces Fréchet Audio Distance by roughly an order of magnitude, from $13.49$ to $1.25$ for AudioLDM. RAG-Audio approaches nearest-neighbor retrieval in identification while remaining generative; its higher FAD is expected because retrieval directly replays real audio. An autoregressive negative control, which lacks an initializable latent trajectory, shows no comparable gain, attributing the improvement to trajectory initialization. These results suggest that retrieval-guided initialization can mitigate prior domination in brain-to-audio generation.

Large audio-language models (LALMs) have demonstrated strong capabilities in understanding diverse audio inputs. This diversity includes low-frequency signals that are inaudible to humans but can still enter the model and influence its generation. However, the practical impact of such low-frequency inputs on LALMs remains largely unexplored. In this paper, we propose Intermittent Low-Frequency Lockout (ILL), an inaudible red teaming method that evaluates this risk using a universal waveform template in a black box setting. ILL uses Sentence Attention Scale Estimation to determine active intervals and Frequency Confusion Transfer to construct a low-frequency waveform with continuous phase from corpus spectral variation. To mitigate this risk, we propose Distributional Requery Guard (DRG) to detect low-frequency distribution shifts and conditionally request a second recording for semantic recovery. Across six LALMs and multiple audio understanding tasks, ILL reduces accuracy by up to 67 percentage points while receiving a mean human audibility rating of 1.33, close to 1.17 for clean audio; DRG raises mean attacked accuracy from 28.5% to 46.1% after clean reacquisition. These findings identify a previously overlooked safety risk for LALMs and provide a foundation for future research on robust audio understanding.

Omni-modal LLMs jointly process audio, video, and text, but long multimodal sequences incur substantial prefill and KV-cache costs. Existing omni-modal compression methods primarily focus on pre-LLM token reduction, leaving modality-specific compression across the LLM boundary underexplored. We propose A-PACK, a two-stage framework that defers audio pruning until query-conditioned multimodal interactions emerge. Our analysis shows that audio exhibits higher task-relevant information density and representational diversity per token than video. We further find that local audio-visual dynamics provide a more effective cue for visual selection than token-wise matching. We therefore preserve audio and compress video with local dynamics before the LLM, then progressively prune low-relevance audio and visual tokens and their KV-cache entries inside the LLM. Across four benchmarks on Qwen2.5-Omni-7B/3B, A-PACK achieves the strongest average performance among the evaluated prior methods while reducing prefill FLOPs by up to 78% and improving decoding throughput by up to 2.21x.

Emotion is central to human communication, and its expression varies across languages. Large audio-language models (LALMs) achieve strong performance on multilingual speech tasks, yet it remains unclear whether they encode emotion through language-specific correlations or language-agnostic representations. We present the first neuron-level interpretability study of this question. We define Multilingual Emotion Neurons (MLENs) as functional units exhibiting stable emotional selectivity and aligned causal effects across languages, and introduce Consistency-Regularized Fusion (CR-Fusion) to identify them. Across four modern LALMs and 12 typologically diverse languages, emotion-sensitive neurons identified independently per language show minimal overlap, and additional monolingual identification data saturates quickly without isolating more transferable units, motivating identification from pooled cross-lingual evidence. Causal interventions demonstrate that MLENs identified by CR-Fusion provide more precise and transferable affective control than monolingual neuron sets in both zero-shot and low-resource settings. Leave-one-out ablations further reveal asymmetric transfer: individual identification languages, including low-resource ones, contribute non-redundant evidence, while several low-resource languages benefit most from the resulting cross-lingual transfer. Together, our findings provide the first causal, neuron-level account of how LALMs encode emotion across languages, and establish multilingual neuron identification as an effective mechanism for understanding cross-lingual affective behavior.

Recent advancements in Speech Large Language Models have demonstrated remarkable capabilities in understanding complex audio tasks. Despite this progress, their long-context inference remains severely bottlenecked by prohibitive KV cache memory demands. Existing text-centric compression methods struggle here, often disrupting speech continuity or discarding crucial semantic cues. To address this, we propose VoxZip, a train-free, two-stage semantic-anchored KV cache compression framework. The first stage uses automatic speech recognition (ASR) transcriptions as explicit semantic anchors to temporally align, compress, and fuse audio tokens, significantly reducing the initial KV cache while elevating token information density. To further improve the compression ratio, the second stage employs a dynamic filtering strategy based on temporally decayed accumulated attention to evict non-essential tokens while mitigating early-token bias. Comprehensive evaluations on Qwen3-Omni across six diverse audio benchmarks demonstrate the superiority of our approach. VoxZip excels in long-audio reasoning and consistently maintains high-fidelity perception on short-form tasks. Notably, it sustains over 90% of the uncompressed baseline performance even under an aggressive 20x KV cache compression in long-context scenarios. Furthermore, at a 4x compression ratio, VoxZip yields a 1.9x increase in inference throughput alongside a 3.3x reduction in peak memory overhead. Code and models will be available at https://github.com/MM-Speech/VoxZip.

Audio dramas weave dialogue, sound effects, and music into immersive stories. Creators often adapt books into audio dramas, but this process remains labor-intensive, requiring them to interpret source material, author scripts, generate audio assets, and assemble them on a timeline. Because story elements like characters and scenes manifest across many interdependent assets, a single change can ripple into manual updates across the entire project. We present Dramarrator, an audio drama authoring tool built around object-based audio editing, where these story elements are represented as editable objects. Dramarrator extracts these objects from a book, generates linked audio assets (speech, sound effects, and music), and composes a multi-track audio drama. Edits to any object (e.g., a character's voice) automatically propagate to all dependent assets. In a user study with professionals (N=8), Dramarrator significantly lowered task load when creating audio dramas. A listener study (N=300) shows that creator-refined output from Dramarrator approaches the quality of productions made with existing professional tools, and an exploratory study (N=3) suggests object-based editing lowers entry barriers and generalizes beyond audio dramas.

The proliferation of AI-generated music in broadcast media raises concerns about transparency and fair compensation, but reliable detection under real broadcast conditions remains unresolved. Existing studies report substantial performance degradation in this domain, yet their evaluations are limited to synthetic broadcast data. To address this gap, we introduce BAMM (Broadcast AI-Music Monitoring), a 40-hour dataset of real-world television recordings containing AI-generated and human-made music. We compare clean-trained and broadcast-trained CNN variants across three progressively more challenging scenarios: Clean Foreground Music (CFM), Synthetic TV Broadcast (STB), and Real TV Broadcast (RTB). Both models achieve near-perfect performance on CFM but degrade substantially under synthetic broadcast conditions. Broadcast-oriented training improves robustness compared with clean training, although performance remains limited. On RTB, evaluated using BAMM, both models degrade further and show substantial score overlap between AI-generated and human-made music. These results expose a critical domain gap and show that current training approaches on CNN-based detectors remain insufficient for reliable AI-generated music detection in broadcast monitoring.

Audio-language models (ALMs) are increasingly used as judges for speech-to-speech systems, but a judge that receives audio may not actually use paralinguistic evidence. We introduce counterfactual audits for paralinguistic response evaluation. Each audit item holds the transcript fixed while varying affect, prosody, or the timing of an affective shift, forcing a valid judge to track the audio cue rather than lexical content or response style. We evaluate ALM judges using a native one-context judgment protocol and a contrastive recoverability control, then further decompose each item into its constituent perception and response-mapping skills. This yields useful diagnostic states that identify different sources of judge failures. Across Gemini, GPT, and open audio models, we find that contrastive success often overstates native judge reliability, and that similar aggregate accuracies can hide different failure modes. These results suggest that ALM judges should not be evaluated by accuracy alone, instead requiring thorough behavioral audits before deployment.

Pansori is a traditional Korean vocal genre whose mode system (jo) is defined not by scale alone but by the entanglement of pitch collection, microtonal ornament (sigimsae), and vocal timbre. In this study, we introduce a 46-hour frame-level pansori mode annotation, expert-labeled across all five canonical batang, and evaluate four complementary input representations (mel spectrogram, F0 contour, MIDI piano roll, and a multi-cultural SSL encoder) under two split strategies designed to detect shortcut learning. Across the three well-represented modes, performance degrades by only 2.1--3.6 points of F1 when entire works are held out, indicating that the models learn mode-relevant features rather than memorizing repertoire. Per-class results further show that source separation removes the percussion cue on which changjo depends, and that generic multi-cultural pre-training fails specifically on the Ujo--Gyemyeonjo distinction. Qualitative analysis of cross-modal disagreement recovers musicologically documented phenomena and agrees with published score-based analyses of modern changjak pansori.

Automatic music mixing aims to combine multitrack recordings into a balanced and coherent musical piece. Because the content of different songs and the subjective preferences of mixing engineers jointly shape the final outcome, a practical system should deliver well-balanced mixes while allowing for controllable stylistic variation. However, most existing methods treat automatic mixing and mixing style control as separate tasks, making it difficult for a single system to produce high-quality mixes while remaining editable and style-aware. To address this limitation, this paper presents Diff2Mix, a generative automatic mixing system based on diffusion models and a differentiable mixing console. This system offers two levels of optional user control: a reference audio enables overall production style control, and the differentiable mixing console provides explicit audio effects parameters for interpretability and fine-grained optimization. We demonstrate our system's competitive performance through both objective and subjective evaluations in terms of mixing quality and control ability. We provide code and audio samples at our project page https://zys711.github.io/Diff2Mix .

While modern 3D reconstruction excels at modeling object geometry and appearance, it largely ignores the rich acoustic cues revealed through physical interaction. Object impact sounds convey material, stiffness, and structural properties that complement vision, yet existing impact sound modeling approaches either rely on expensive physics-based simulation or require large datasets to generalize in a purely data-driven manner. We introduce Audio-Visual Modal Sound Field (AV-MSF), a novel object-level acoustic representation reconstructed from multi-view images and only a few impact sound recordings. AV-MSF builds on 3D Gaussian Splatting integrated with dense 3D visual feature to provide a strong geometry-aware prior, and represents the impact sound field using compact, physically meaningful modal parameters, enabling robust few-shot reconstruction. Experiments on two real-world datasets show that AV-MSF achieves state-of-the-art impact sound rendering, outperforming both physics-based and data-driven baselines. Furthermore, we demonstrate downstream applications enabled by our representation, including contact localization and object sound editing.

Spoken Language Understanding (SLU) is the core component of task-oriented dialogue systems and a pivotal link in achieving seamless human-agent interaction. While traditional SLU can effectively extract user semantics for closed-set tasks after in-domain supervised fine-tuning, it faces significant challenges in leveraging in-context learning for open-domain tasks due to its ambiguous rule definitions. This work proposes Spoken Function Calling (SFC), a novel semantic understanding perspective that optimizes semantic understanding with structured rule definitions, to evolve beyond traditional closed-set SLU. Specifically, we curate and extend a suite of spoken functions based on traditional SLU datasets, construct a multi-agent system to synthesize the SFC-Bench dataset, evaluate the performance of Large Language Models (LLMs) and Large Audio Language Models (LALMs), and enhance the SFC capabilities of LALMs through post-training. Experiments demonstrate that SFC outperforms traditional SLU, substantially enhancing the semantic extraction accuracy for LLMs and LALMs.

Video-to-audio (V2A) generation extends image-to-audio generation (I2A) by introducing consecutive frames that provide essential temporal cues for audio synthesis. However, existing conditional diffusion-based V2A methods typically enhance visual conditioning with additional audio-visual supervision, acoustic structure prediction, or reasoning from large multimodal models, requiring extra networks or strong inductive biases. Inspired by recent advances in visual representation learning, we introduce TD-V2A, which leverages temporal differences (TD) as the key representation that distinguishes V2A from I2A, enriching visual conditioning with minimal architectural modification. We first investigate TD at both the frame and feature levels to identify the most effective representation level at which TD complements visual representations. Based on these findings, we develop a hierarchically continual learning strategy and an annealed temporal differences guidance method to progressively learn and exploit TD information during diffusion training and sampling process, respectively. Extensive experiments on benchmark datasets demonstrate that effectively exploiting TD through our proposed framework significantly improves end-to-end V2A generation quality, even outperforming dedicated V2A representations such as contrastive audio-visual pretraining.

Collaborative music agents need internal representations rich enough to support both understanding and generation, yet flexible enough for a workflow where the human retains agency. We present a hierarchical self-supervised ``world model'' for symbolic music: a 2.55M-parameter Swin V2 encoder trained on MIDI piano-roll images with JEPA-style objectives (pitch- and time-shift equivariance, masked embedding prediction, and a distributional regularizer), using no labels and no music-theory vocabulary. Probing the frozen embeddings shows that the level at which a musical property becomes decodable tracks its musical time scale: phrase boundaries are read off the coarsest levels, note density and harmonic detail off the finest. Temporal and phrase structure emerge from the self-supervised objectives alone, while harmonic content must be asked for; a small chord-supervision head raises joint chord recovery from .18 to .54, and key detection, which is never supervised, from .16 to .70. Following the Representation AutoEncoder paradigm, a conditional flow-matching model stands in for a trained decoder, flowing in pixel space from PCA-reduced conditioning: it reproduces a target window at pixel F1 $0.996$, and the same per-level conditioning dropout that controls how far variations stray also enables graphical prompting for masked inpainting with no inpainting-specific sampler. The pipeline runs on CPU producing a suggestion in $2.8$ s, or $0.6$ s on Apple MPS, which we demonstrate in a live interactive demo. In concert with an LLM-based brain, these capabilities supply the core of a collaborative music creation agent in service of, rather than in place of, human agency.

Large Audio-Language Models (LALMs) excel at general speech understanding; however, adapting them to fine-grained tasks like Speech Emotion Recognition (SER) remains a significant bottleneck. Current Parameter-Efficient Fine-Tuning (PEFT) methods typically operate in flat Euclidean space, and this geometry fails to capture the multi-granularity nature of emotion cues, which range from low-level prosody to high-level semantics. To address this, we propose HyPASE, a hyperbolic PEFT framework for LALM-based SER. HyPASE leverages the Poincare ball model, using the hyperbolic radius as an explicit proxy for representational granularity. The framework consists of two core components: a Hyperbolic Geometric Adapter (HGA) for layer-adaptive weight modulation, and an Emotion-aware Multi-capacity Cross-modal Aggregator (EMCA) that compresses multi-scale features into compact audio prefixes. Empirical results on standard benchmarks show that HyPASE outperforms Euclidean PEFT baselines across all metrics on MELD and achieves a notable Unweighted Accuracy gain on IEMOCAP, particularly in class-imbalanced emotion recognition, with the accompanying slight Weighted Accuracy trade-off reflecting hyperbolic space's geometric prioritization of minority-class representations; furthermore, HyPASE achieves robust zero-shot cross-dataset generalization within a constrained parameter budget. By grounding the adaptation process in hyperbolic geometry, HyPASE offers a highly efficient path for LALM fine-tuning.

We investigate an unobtrusive and $24\times7$ human distress detection and signaling system, Always Alert, that requires the smartphone, and not its human owner, to be on alert. The system leverages the microphone sensor, at least one of which is available on every phone, and assumes the availability of a data network. We propose a novel two-stage supervised learning framework, using support vector machines (SVMs), that executes on a user's smartphone and monitors natural vocal expressions of fear---screaming and crying in our study---when a human being is in harm's way. The challenge is to achieve a high distress detection rate while ensuring that the false alarm rate is a manageable overhead, while a typical smartphone user goes about living life as usual. We train the learning framework with carefully selected audio fingerprints of distress and of varied environmental contexts. The audio is used to tune the learning framework to obtain a desirable distress detection rate and false alarm rate (FAR). The ability of the proposed framework to detect distress in rather challenging audio environments is demonstrated. Exploiting the time contiguous nature of false alarms further allows us to reduce the FAR. We show the feasibility of using our framework anytime and anywhere by testing it over many hours of audio fingerprints recorded by volunteers on their smartphones, as they went about their daily routines. We are able to achieve high distress detection rates at an average overhead that is equivalent to about 1 facebook post every 3 to 4 hours.

Passive acoustic monitoring is an important tool for biodiversity assessment and wildlife conservation because it supports continuous and non-invasive monitoring of species across large spatial and temporal scales. Robust monitoring remains challenging because many datasets contain sparse positive labels, where species presences may be confirmed while unannotated species cannot be assumed absent. In this work, we study transfer learning under sparse positive labels using BirdCLEF+ 2026 as a target benchmark and BirdCLEF 2021, iNatSounds, WABAD, and BirdSet as external bioacoustic sources. We introduce a multi-source reliability framework that models heterogeneous bioacoustic datasets as distinct supervision sources with differing reliability. Our approach achieves 0.584 macro average precision and 0.860 macro AUC on public BirdCLEF+ 2026 validation labels while outperforming naive source pooling strategies. The strongest gains arise from passive acoustic monitoring datasets and biologically informed source selection. Our findings suggest that transfer learning in bioacoustics is fundamentally a weak supervision and negative transfer problem.

Humans recognize a musical passage even when it is shifted in time or transposed in pitch, indicating a notion of equivariance in the representation space. Our analysis, however, shows that standard music transformers map such time-shifted or pitch-transposed inputs onto uncorrelated representations: these models become progressively less equivariant as they scale in size or train longer. This suggests that in standard music transformers, additional model capacity is allocated to memorizing absolute patterns rather than capturing shared musical structures. In this paper, we propose the Equivariant Music Transformer (EMT), which enforces equivariance through self-distillation by jointly optimizing a next-token-prediction and an auxiliary equivariance regularization loss. We find that the additional equivariance loss acts as a beneficial regularizer, simultaneously improving next-token prediction and producing equivariant latent representations. Through both objective and subjective evaluations, EMT demonstrates superior equivariance and generative capability compared to data augmentation, feature engineering, and state-of-the-art (SOTA) baselines. More broadly, our findings reveal that standard language modeling methods alone do not capture music's translational symmetries, and dedicated inductive biases are required to produce better music representations. The code, weights and demos are available online.

Recent audio restoration increasingly relies on large-scale conditional latent generative modeling, including diffusion, Schrodinger Bridges, and Flow Matching variants, to invert degradations such as bandwidth limitation or noise. We present an analysis of the performance of various state-of-the-art methods compared to simple arithmetic transformations in the latent spaces of multiple neural codecs for musical bandwidth extension. We show that estimating a single transport vector between the clean and degraded latent centroids on a reference set, and adding it to degraded latents, can yield restoration performance competitive with large diffusion models. This suggests, first, that some neural codec latent spaces exhibit structure aligned with audio bandwidth; and second, that in such cases complex conditional models may offer only limited gains over a simple vector addition. We argue that these findings reveal an interesting avenue for future research whereby models could take advantage of the latent space structure in order to offer greater training and parameter efficiency, and overall better performance. Additionally, we propose to consider this simple arithmetic transformation as a baseline for music bandwidth extension research, as it allows an assessment of the contribution of learnable parameters towards restoration performance.

What is music style? Though often described using text labels such as "swing," "classical," or "emotional," the real style remains implicit and hidden in concrete music examples. In this paper, we introduce a cross-modal framework that learns implicit music styles from raw audio and applies them to symbolic music generation. Inspired by BLIP-2, our model leverages a Querying Transformer (Q-Former) to extract style representations from a large, pre-trained audio language model (LM), and further applies them to condition a symbolic LM for generating piano arrangements. We adopt a two-stage training strategy: contrastive learning to align auditory style with symbolic expression, followed by generative modeling for music arrangement. Our model generates piano performances jointly conditioned on a lead sheet (content) and a reference audio example (style), enabling controllable and stylistically faithful arrangement. Experiments demonstrate the effectiveness of our approach in piano cover generation, style transfer, and audio-to-MIDI retrieval, achieving substantial improvements in style-aware alignment and music quality.

Audio reasoning is essential for machine understanding of the acoustic world. Reinforcement learning with verifiable rewards can elicit such reasoning, yet existing reward designs are complementary in their limitations: outcome-based rewards supervise only the final answer and let the model reach it without attending to the audio, whereas process-based rewards score the reasoning itself but rely on coarse, hand-crafted, and fixed criteria that neither adapt to each question nor stay grounded in the acoustic evidence. Moreover, questions differ in what they demand, with some hinging on perception and others on multi-step reasoning, and any static criterion weakens as the policy improves. Supervising the reasoning process with fine-grained, audio-grounded, and adaptive rewards is therefore crucial, yet challenging since such rewards are impractical to design by hand for every sample. To this end, we introduce AudioRubrics, a reinforcement learning framework that supervises audio reasoning with self-evolving, audio-grounded rubric rewards. AudioRubrics synthesizes per-sample rubrics from the raw waveform and, conditioned on the model's own rollouts, regenerates and reweights criteria per group, supplying a continuous learning signal that keeps targeting the current policy's weaknesses as static criteria saturate. Comprehensive evaluations across three audio reasoning benchmarks reveal that AudioRubrics substantially outperforms a wide range of open-source and training-based baselines. Furthermore, our analysis shows that the gains scale with the capability of the rubric generator and judge, and AudioRubrics converges to a stable reasoning length that avoids both degenerate collapse and unbounded growth. The improvement in audio perception further demonstrates the effectiveness of anchoring supervision in the acoustic evidence. Our project page is available at https://audiorubrics.github.io.

Artificial intelligence offers substantial potential for acoustic monitoring of animals, from welfare assessment in precision livestock farming to wildlife conservation and ecological research, where vocalizations can indicate health, stress, and social states earlier and at lower cost than manual observation. However, recordings in these settings are obtained under uncontrolled conditions, including environmental noise, reverberation, overlapping calls, and sensors that degrade without notice. As a consequence, automated classification of animal vocalizations remains challenging, and the two dominant acoustic representations show complementary limitations: raw waveforms preserve temporal microstructure but degrade under clipping and reverberation, while log-Mel spectrograms capture harmonic organization but lose phase information and are sensitive to broadband noise. To address these challenges, we propose Uncertainty-Aware Fusion (UAF), a dual-stream framework that estimates Gaussian uncertainty for each representation and fuses them via uncertainty weighting. This mechanism assigns greater weight to the more confident representation with no reliability labels required. In a cross-species, identity-based evaluation excluding all individuals seen during training, UAF (mean pooling) achieves 59.4% accuracy / 39.7% macro F1 on the 17-class SoundWel pig vocalization benchmark and 73.1% accuracy / 71.5% macro F1 on the 3-class DogBark dataset, outperforming static-concatenation fusion by 15.7% and 20.4% relative macro F1, respectively. Ablations over four temporal aggregation strategies show that uncertainty fusion, rather than the temporal characteristics of animal calls, is the primary driver of the performance gain.

Music restoration seeks to recover a clean signal from an observed recording degraded by an unknown effect, distortion, or corruption. Existing systems often rely on paired training data and distortion-specific supervision, which limits their use when the forward process is not known in advance. We propose LOUDAR (Latent-space Optimization of Unknown Distortion for Audio Restoration) a general-purpose restoration method that operates in the latent space of a pretrained audio autoencoder and models the unknown distortion as a learnable latent operator. At inference time, LOUDAR alternates between estimating the clean latent variable and updating the latent operator parameters. An unconditional latent diffusion model provides a prior over clean audio and regularizes this inference by steering the latent estimate toward the manifold of clean recordings. Because the degradation model is adapted per input, the approach is broadly applicable across diverse restoration problems. We evaluate LOUDAR on singing voice effect removal and restoration, as well as guitar distortion removal, and show that it consistently improves over degraded inputs and is competitive with supervised and unsupervised baselines in waveform and latent domains.

Complex acoustic problems may require models to perform acoustic operations, interact with external tools and reason over the resulting textual or processed-audio observations rather than answer directly from a fixed audio input. We study such problems as tool-interactive audio reasoning and develop SpeechAgent-R, an audio agent that coordinates its intrinsic multimodal understanding with external skills and tools. To support this capability, we construct HIU-Corpus, comprising 65,492 interaction trajectories and 507.6 hours of audio across 24 tasks, 8 skills and 9 tools. SpeechAgent-R first learns structured interaction behaviors through trajectory-based supervised fine-tuning and then improves its decisions through multi-turn reinforcement learning. We further introduce HIU-Bench to jointly evaluate task performance, interaction quality and generalization to diverse task settings. It contains 1,395 samples across 56 tasks, including in-distribution (ID) and out-of-distribution (OOD) splits with substantial shifts in tool usage and workflow composition. SpeechAgent-R achieves 84.17 on ID tasks and 70.94 on OOD tasks, improving over the base model under the same agent harness by 15.40 and 14.23 points. These results demonstrate that learning skill and tool coordination improves audio agents' ability to handle diverse task settings and adaptive tool interactions.

This paper describes the participation of team "Go-To-Germany" in the ImageCLEF 2026 Audio Deepfake Detection and Generation task. Our detection system, built on a four-backbone self-supervised learning (SSL) ensemble combining WavLM-Large, Wav2Vec2-XLS-R-300M, ECAPA-TDNN, and x-vector representations, achieved a final score of 0.9522 on the official ImageCLEF 2026 evaluation, with perfect accuracy (1.0000) on participant-generated deepfakes and 0.8875 on the held-out organizer ground-truth real data. For the Generation sub-task, our official team submission, an F5-TTS v1 baseline processed with a uniform reverberation pass and submitted as a deliberate anti-forensic probe, ranked first with a final score of 0.4304 (word error rate (WER) 4.99%, character error rate (CER) 2.07%); details of our four-model program (GLM-TTS, F5-TTS, XTTS v2, CosyVoice3), from which the official entry was drawn, appear in the paper. We present a cross-track analysis revealing a pronounced asymmetry: our detection system identifies 100% of participant-generated deepfakes, while our official generation entry, despite ranking first in the Audio Generation sub-task and evading 61.4% and 56.2% of participant and organizer detectors, attains a Final Score of 0.4304 against 0.9522 on the Detection side. We further report falsification-based ablation experiments (LOSO 56-speaker cross-validation, three-region backbone geometry, bootstrap confidence intervals, and PCA analysis) that motivate our architectural-insurance hypothesis for multi-backbone SSL ensembling. We complement these results with five cross-track insights and five pre-registered falsification experiments connecting generation-side evasion to detection-side design decisions, and we openly report an 11.25% false-positive gap on held-out organizer real recordings as the principal open challenge for deployment.

Scaling the corpus is the default remedy when a contrastive representation lacks an attribute. We report a case where it does nothing, and identify what does: adding a lexical-speech round to a frozen-base multimodal embedding model raises zero-shot keyword spotting by 76 points while reducing speech-emotion recognition by 14. The loss is not a capacity limit: fine-tuning on 7,442 clips from a prosody-controlled corpus recovers emotion past its pre-speech level at a five-point keyword cost. Nor is it data volume: 29,428 mined clips whose captions explicitly name emotions, at matched exposure, move emotion by -0.0007. The difference is structural: a contrastive objective encodes an attribute only when the in-batch negatives cannot be separated without it; the controlled corpus holds sentence content fixed, so prosody is the only separating signal, whereas mined captions name emotion yet remain separable by scene content. Intervention on the same audio confirms causality: raising caption similarity does not recover emotion, but collapsing caption diversity so that emotion becomes the only separating axis recovers it by 8.9 points across three seeds, with a smaller, same-signed gain on a non-acted corpus, while keyword accuracy trades back. Corpus structure, not size or caption vocabulary, controls what a contrastive audio embedding encodes.

Short segments of perceived speech can be retrieved from non-invasive magnetoencephalographic (MEG) recordings by deep networks trained with a CLIP-style objective against wav2vec 2.0 audio embeddings. Yet their weights do not map onto electrophysiological quantities, and it remains unclear which speech properties drive retrieval. We build on a high-performing MEG-to-audio retrieval architecture but redesign both its front end and decoder. Its spatial attention operates on a flattened sensor layout; we replace it with spherical harmonics defined on the three-dimensional MEG helmet geometry. We reduce the subject-specific representation from 270 to 25 branches, add a temporal filter to each branch to match it to a neuronal source in space and time, and make the convolutional decoder shallower. Ocular and cardiac components are removed before training to reduce the risk of stimulus-locked shortcuts. On MEG-MASC, the model reaches 39.75 +/- 0.34% Top-1 accuracy among 1005 candidates across six trained solutions, with about 20 times fewer decoder parameters. Its weights map to source space, recovering generators consistent with the speech-perception network, while left-lateralized branches carry higher-frequency rhythmic components not evident on the right. Paired MEG occlusion shows that 15 of 19 stimulus features contribute, with the largest effects for silence, sound intensity, vowels, and acoustic onsets. Random word lists behave oppositely: substituting narrative MEG into them improves retrieval, indicating that activity without narrative structure carries less recoverable information than activity during coherent speech. The wav2vec target can be reduced to about twelve learned feature dimensions without loss of accuracy, whereas strong temporal compression causes a clear loss. Together, source mapping and input interventions reveal what drives retrieval.

Speech editing for content creation requires precise control over both what an edit should do and where it should apply. Free-form natural language provides a flexible interface for expressing edit requests, but its ambiguity may leave the intended operation, parameters, or target region underspecified. We study a precise and explicit interface for speech editing: a transcript-grounded structural edit instruction with XML-style tags explicitly specifies typed operations and localizes them to transcript spans or boundaries. This semantic timeline avoids explicit timestamp alignment and provides an externally inspectable contract for compositional edits. We instantiate the interface in dots$.$tts$.$edit, an editor adapted from the continuous autoregressive dots$.$tts foundation model. Four representative speech-creation controls cover lexical content, affective expression, pitch and speaking-rate delivery, and temporal phrasing through text, emotion, prosody, and pause editing. Task-specific data pipelines construct operation- and scope-controlled pairs while retaining source-derived context outside each target region. We further introduce doteBench, a bilingual evaluation suite that measures precise instruction following, local preservation, and audio quality across the four controls and their composition. Experiments show leading overall instruction following and local preservation across its five editing categories, while audio quality remains comparable to existing open-source systems. Across three Seed-TTS-Eval shards, the model shows negligible differences from the base model in zero-shot TTS recognition error rate and speaker similarity.

High-polyphony symbolic music is increasingly used in generation, analysis, and arrangement, yet many downstream tasks require bounded representations with fixed tracks or slots. Converting richly orchestrated scores into compact forms is therefore necessary, but existing approaches relying on heuristic simplification or generic representation-space reduction often fail to preserve structural roles, orchestration compatibility, and playability under strict budgets. To address the issue, this study reformulates the compression problem as a fixed-budget structured routing problem and proposes Unbalanced Optimal Transport for Information Routing (UOT-IR), a training-free framework based on constrained unbalanced optimal transport. UOT-IR combines an orchestration prior, adaptive marginal relaxation, temporal decoding, and playability-aware projection to produce compact and musically coherent bounded representations. This work further studies two practical settings under the same slot budget: template standardization, which maps each input to a predefined bounded template, and adaptive preservation, which retains representative content without assuming an external template. Experiments on the SymphonyNet corpus show that UOT-IR delivers strong overall performance across both settings, including the best Note-F1 in adaptive preservation (0.9120), together with the lowest structural cost (14.7165) and bad structural confusion rate (0.3406) in template standardization. This work establishes a principled paradigm for fixed-budget symbolic music compression, offering a practical path toward compact, structured, and musically coherent symbolic representations.

All-type audio deepfake detection requires authenticity decisions across speech, environmental sound, singing voice, and music, while the audio type is unavailable at inference time. In AT-ADD Track2, this setting creates a hidden audio-domain condition: the binary real/fake label is shared across domains, but representation structure and detector-score behavior vary with audio type. We present a closed-condition routed system that first recovers the hidden audio domain and then interprets detector scores within the selected branch. The AudioType-BEATs-6s Router estimates audio type from a 6-second window; speech inputs are handled by the Speech-XLSR Expert, while sound, singing, and music rely on EAT-based general-audio experts with branch-local score interpretation. Development-set representation analysis, router-family comparisons, and component results show audio-domain separation and complementary detector strengths across audio types. On the official AT-ADD Track2 final evaluation, the system achieves 96.10% Track2 Macro-F1 and ranks first on the final leaderboard, with type-wise Macro-F1 scores of 88.07%, 98.18%, 99.07%, and 99.08% for speech, sound, singing, and music, respectively. These results support recovering the hidden audio domain before interpreting detector scores in all-type audio deepfake detection.

Aerial robots generate structured aeroacoustic fields during flight, yet these signals have been underexplored as a source of onboard relative perception, particularly under the strong ego-acoustic interference generated during simultaneous flight in various outdoor conditions. We introduce embodied passive aeroacoustic perception, a sensing paradigm in which an aerial robot infers actionable relative-state information from the naturally generated sound of flight while operating within its own evolving aeroacoustic field. We present SonicFly, a passive aeroacoustic perception framework that enables one unmanned aerial vehicle to estimate and follow another using only the leader's intrinsic flight sound, without active acoustic signaling, inter-robot communication, GPS sharing, or external sensing infrastructure. The system uses a lightweight four-microphone array, rotorcraft-informed acoustic representations, a neural bearing-range estimator, and confidence-gated filtering for closed-loop flight. Through acoustic characterization, onboard localization, and outdoor pursuit experiments, we show that multirotor aeroacoustic signals contain sufficient information to support relative perception despite strong ego-acoustic interference, environmental variability, and changing flight geometry. During acoustic-only pursuit, SonicFly achieved a mean distance-maintenance error of 1.34 m across diverse outdoor trajectories and operating conditions. Analysis of the acoustic channel reveals design principles governing embodied passive aeroacoustic perception, including the roles of harmonic structure, spectral separability, and spatial acoustic cues in determining observability. Our results establish the feasibility of embodied passive aeroacoustic perception for aerial robots and suggest that naturally generated behavioral signals can serve as information for robotic perception and coordination.

Unified audio models capable of audio understanding, audio generation and, increasingly, audio editing are proliferating rapidly. Yet a basic question about them remains unanswered: do the two heads of a unified model agree about the same audio? Current practice evaluates each capability in isolation on specialized benchmarks, and never asks whether a model can make sense of its own generations. We present TORUS, the first self-coherence test for audio-native unified models. TORUS comprises 48 three-stage self-coherence tests carrying 432 six-option questions spanning speech, sound and music across five task families. We holistically evaluate five open unified models alongside a Cascaded Baseline that combines state-of-the-art specialized generation, editing and understanding models. The best unified model answers 50.5% of questions against the Cascaded Baseline's 63.2% and a 16.7% chance floor. Models struggle on audio editing. Among the evaluated audio models (specialized and unified), we observe limited self-coherence, and thus position self-coherence as an essential test for future audio systems.

Speech deepfake detection has expanded in scope with increasingly heterogeneous spoofing mechanisms, including speech synthesis, voice conversion, vocoder reconstruction, and neural-codec resynthesis. The resulting spoofing artifacts can be further shaped by variability in source speech, recording environments, and transmission channels. This variability makes robust generalization across heterogeneous conditions a central requirement for practical detection systems. This report presents Teffic-Audio, a general speech deepfake detection system designed for comprehensive evaluation environment. Teffic-Audio adopts a straightforward detector architecture consisting of a Conformer-based speech encoder, multi-head attentive statistics pooling, and a binary classifier. Rather than relying on additional architectural complexity, the system improves generalization through its training recipe, which integrates multi-source data, attack- and source-balanced sampling, and diverse audio augmentation. Trained only with open-source data, Teffic-Audio achieves a pooled EER of 1.454% on the 14 test sets of Speech-DF-Arena, outperforming all currently public systems on the leaderboard. It also obtains the lowest EER on five individual test sets and shows a favorable performance-complexity trade-off compared with larger leading systems. Overall, Teffic-Audio provides a strong and practical reference system for general speech deepfake detection.

Existing singing voice synthesis systems often require predefined durations, explicit duration prediction, or time-aligned acoustic guidance, which limits their compatibility with practical composition workflows. We propose VocalRender, a score-native system that directly synthesizes singing from lyrics, pitches, symbolic note values, and tempo. It uses an interleaved lyric--note representation and an autoregressive diffusion model to generate continuous acoustic latents while predicting the output length, eliminating the need for explicit duration prediction. Trained on a 2,300-hour singing dataset, VocalRender achieves strong intelligibility, strong melody control, and high speaker similarity across both in-domain and out-of-domain benchmarks. Notably, it outperforms the strongest baseline by $0.42$ points in naturalness CMOS, demonstrating the effectiveness of our proposed score-native architecture.

AI music generators leave predictable spectral artifacts determined by their architecture. Existing detectors exploit these artifacts with near-perfect accuracy on raw generated tracks, but their performance collapses under simple audio manipulations, such as speed modification or pitch shifting. We address this open robustness problem by introducing a frequency-scaling-invariant detection pipeline that aims to prevent this kind of attack by design. Our method maps audio onto a log-frequency axis via a log-STFT remapping. A single learned cross-correlation filter, combined with max-pooling, provides shift invariance at inference time. Training uses a hybrid loss that jointly supervises binary detection and artifact-peak localization, regularizing boundary weights. Because robustness to speed change is built in by design, the detector is also interpretable: it outputs both a binary decision and an estimate of the applied speed-change factor.

With the development of audio large language models (AudioLLMs), audio captioning needs to move from brief descriptions toward open-ended and fine-grained free-form descriptions. Existing evaluations often focus on generation quality or task performance, making it difficult to diagnose information coverage and description reliability. We propose MMAC, a \textbf{M}assive \textbf{M}ulti-dimensional benchmark for \textbf{A}udio \textbf{C}aptioning. MMAC contains 5,638 audio clips from more than 20 data sources, covering 6 capability categories and 15 evaluation dimensions. Given a model-generated caption, MMAC checks whether it mentions relevant information in the target dimension and whether the mentioned content is consistent with the reference label. We evaluate representative open-source and proprietary AudioLLMs. Results show clear differences across evaluation dimensions, information coverage, and description reliability. We will release the MMAC benchmark and evaluation code.

Audio-video generative models achieve impressive quality but suffer from high latency, making them unsuitable for real-time applications. Although several streaming audio-video generation methods have been proposed, they remain costly and fail to support long-form generation. To address this, we propose \textbf{Ripple}, a real-time joint audio-video generation system with a cross-modal recurrent memory mechanism. To enable efficient streaming inference while preserving long-term context, Ripple combines a fixed-length sliding-window attention with modality-specific memory states that continuously summarize audio and video context. Cross-modal memory interaction is further introduced to enhance audio-visual synchronization. To learn this memory-augmented model effectively, we devise a three-stage training recipe: (1) adapting a bidirectional audio-video teacher to block-wise causal attention with simulated memory, (2) optimizing the memory construction and interaction pipeline through end-to-end distillation, and (3) applying online reinforcement post-training tailored for streaming audio-video generation. As a result, Ripple achieves ~28 FPS at 480P resolution, over faster than the teacher, while capable of coherent long-form generation. Extensive experiments on both short-video and long-video benchmarks demonstrate our superior performance over existing offline and online joint audio-video generation methods.

Existing audio forgery detection and localization (AFDL) methods often overfit dataset-specific low-level artifacts, limiting their generalization to subtle, localized, and unseen manipulations. Recent audio large language model (ALLM)-based approaches cast AFDL as question answering but still model forensic evidence implicitly, without linking manipulation cues to predictions. To bridge this gap, we propose ThinkOmni, a reasoning-driven omni-modal large language model that jointly performs explicit forensic reasoning, spoofing detection, and temporal manipulation localization. To enable explicit reasoning supervision, we construct Forensic-Aware Chain-of-Thought (FACoT), a 100K-sample dataset with structured forensic evidence and reasoning annotations. Leveraging FACoT, we introduce Forensic-Aware Modality-Incremental Learning (FMIL), which progressively aligns semantic, acoustic, and spectral-visual representations with the LLM backbone to capture complementary forensic cues. We further propose Forensic-Consistent Multi-task Loss (FCML), which combines weighted cross-entropy with an adaptive localization loss to coordinate reasoning generation, spoofing detection, and temporal localization. Extensive experiments show that ThinkOmni achieves strong cross-dataset generalization in both detection and localization. Code, models, data, and inference examples are available at https://beyond0814.github.io/ThinkOmni/.

Audio-capable foundation models enable end-to-end spoken interaction, but they also introduce safety risks beyond transcript content. It remains unclear how much jailbreak capability can arise from matched-text variation in speech delivery rather than from lexical rewriting or broader style transfer. We study this question by holding transcript content fixed and varying six speech-delivery presets whose acoustic attributes may co-vary. We present PJ-Break, a black-box evaluation protocol with presets targeting arousal, authority, and speaking rate, together with AdvAudio-Prosody, a 600-sample benchmark with acoustically verified attributes. On the exact post-QC Qwen2-Audio panel, the Q=1 Panic (38/95), Anger (35/95), and Fast (32/95) presets are all well above Neutral (4/95). The fixed six-query pool covers 44/95 Qwen2-Audio seeds and 15/95 GPT-4o seeds and exceeds a matched-budget StyleBreak reimplementation (27/95) on Qwen2-Audio. A same-voice pool excluding the confounded Commanding condition still reaches 40/95, and a retained-panel ablation shows emotional-delivery audio alone (44/95) is far more effective than emotional text alone (11/95). Exploratory surrogate diagnostics and pilot mitigation observations are secondary, non-core analyses. Overall, matched-text speech delivery should be treated as a first-class factor in Audio LLM safety evaluation

Large language models (LLMs) are increasingly embedded in everyday life and widely used for information seeking, raising concerns about their potential to perpetuate social biases and reinforce stereotypes. In this study, we investigate gender bias in LLMs through the lens of their associations with musical instruments. Building on social-science research on the cultural gender-typing of instruments, we introduce Symphony-Bias, a parallel multimodal dataset spanning text, vision, and audio. We evaluate ten multimodal models with diverse architectures and scales across 22 musical instruments, analyzing how they associate each instrument with three gender categories: {male, female, non-binary}, across three modalities: {text, vision, audio}. Our results show that 92% of instrument-level outcomes align with prior social-science findings, with the harp and drums showing particularly consistent gendered associations across all evaluated models and modalities. We further find that alignment with social stereotypes is weakest in audio, stronger in vision, and strongest in text, suggesting that modality-specific representations can differentially amplify gendered associations with musical instruments.\footnote{The Symphony-Bias dataset will be publicly released upon acceptance of the paper.}

Open Sound Control (OSC) is the dominant wire protocol for real-time parametric control in professional audio, live performance, and virtual production. Large language models can emit plausible OSC, but they hallucinate addresses, mishandle type tags, and fail under paraphrase- unacceptable in show-critical contexts. We present LLM4OSC, a local-first architecture in which models propose structured intent JSON over a human-reviewed device profile, and deterministic code validates, clamps, and encodes before any UDP send. We introduce a frozen evaluation harness with CI gates on wrong-send rate: mismatches that would still pass validation and transmit. On a Max/MSP hero profile (12 patterns; 8 literal + 8 paraphrase + 4 refusal cases), after profile tag enrichment, symbolic slot fill, NL refine, and a retrieval confidence gate, backends B0--B3 all pass frozen gates (100% semantic accuracy, 0% wrong-send). B0 (rules) remains the production default at ~0.05ms; LLM backends remain ~3-4s. Historical few-shot B2 accuracy of 62.5% rises to 100% on this suite only after symbolic post-processing- not because the 0.5B model alone becomes show-safe. We argue for propose-validate-send and wrong-send rate as first-class metrics for language-to-control systems.

Large audio-language models (LALMs) convey acoustic evidence to language decoders through native audio tokens, yet the internal roles of these tokens remain poorly understood. Using temporal audio grounding as a diagnostic setting, we examine how language-model fine-tuning affects the layerwise semantics, decoder accessibility, and temporal output alignment of native audio-token states through four complementary analyses: query-conditioned token semantics, calibrated token readout, temporal-window probes, and residual-delta erasure during generation. Alongside substantial improvements in temporal localization, semantic analysis of Qwen2.5-Omni shows that latent evidence for queried events is already present before fine-tuning and that the audio tokens most strongly aligned with the queried event appear at similar temporal positions before and after fine-tuning. After fine-tuning, event-related information in audio tokens becomes more accessible to the decoder, especially in early and middle layers, and a cross-checkpoint control shows that this improvement arises primarily from decoder adaptation. Temporal probes show that the base checkpoint already contains recoverable information about annotated windows and that fine-tuning mainly improves alignment with each checkpoint's own predicted temporal support. Residual-delta erasure further shows that removing audio-token updates within predicted windows harms timestamp generation more than removing the same number of randomly selected updates. The same broad improvements in decoder readability and prediction alignment also appear in Qwen2-Audio. Together, these results support a semantics-to-readout account in which grounding fine-tuning helps the decoder read existing event evidence and connect it more reliably to temporal outputs.

Latent Acoustic Mapping (LAM) is a self-supervised learning method that generates high-resolution spherical acoustic maps from multichannel recordings without labelled data, matching supervised baselines on direction-of-arrival benchmarks. However, LAM degrades significantly with sparse 4-channel arrays, as the low-resolution cross-spectral matrix captures far less spatial information than the 32-channel inputs LAM was designed for. We benchmark a diverse set of upsampling architectures, spanning lightweight convolutional networks, iterative back-projection models, physics-informed networks, and generative adversarial approaches. We also study whether aligning these upsamplers with LAM by training them jointly or in different stages helps preserve the spatial structure that LAM depends on. Results show that the original full-resolution LAM is the strongest, that separately trained lightweight models are the most competitive learned approaches, and that representation alignment between the upsampler and LAM matters more than model complexity.

Production-ready audio-driven avatar generation requires efficient inference without sacrificing fidelity or motion expressiveness. However, existing acceleration methods often compromise quality through restrictive architectural choices, such as causal attention and short temporal horizons, or by reducing model capacity and resolution. Without such compromises, we propose AptAvatar, a 14B-parameter long-form audio-driven avatar generation framework that delivers fast and expressive inference. For efficiency in production-level applications, AptAvatar addresses the extreme two-step generation challenge. To bridge the gap between the multi-step teacher model and the two-step student model, we introduce Endpoint-Anchored Distribution Distillation. It augments vanilla distribution matching with a dedicated Anchor Score Estimator trained on the trajectory-endpoint distribution defined from a frozen pretrained 4-step bridge generator. This provides an attainable endpoint-level anchor for the evolving two-step student. To improve long-horizon consistency, we further introduce Self-Generated History Replay, which reuses cached outputs from earlier generator checkpoints as history conditions during chunk-wise training. This approximates inference-time conditioning on self-generated histories without costly online rollouts, mitigating quality degradation from accumulated history errors. Extensive experiments demonstrate that AptAvatar generates vivid 720p long-form avatar videos with only 2 NFEs, achieving a 60x speedup while preserving visual fidelity and long-horizon identity. Code is available at https://github.com/TaoLiveAIGC/AptAvatar

Recent generative models are moving beyond silent video or standalone audio synthesis toward the joint generation of synchronized audio and video. Despite this progress, jointly generating audio and video with fine-grained cross-modal correspondence remains challenging due to their fundamental structural differences. Most existing methods use audio and video VAEs trained separately. As a result, the two latent spaces lack cross-modal alignment, leaving the downstream generative model to learn cross-modal synchronization from scratch. We present OmniVAE, a jointly trained audio-video VAE that learns fine-grained semantic alignment between audio and video latent representations. Beyond reconstruction, OmniVAE uses a segment-level audio-video contrastive objective to capture temporal-semantic correspondence and align the two latent spaces. In parallel, it distills features from pretrained modality-specific semantic encoders into each modality, improving the downstream learnability of both latent spaces. Extensive experiments show that both objectives consistently improve the learnability of the latent spaces, translating into higher generation quality and more accurate cross-modal synchronization in downstream text-to-audio-video generation. These findings underscore the importance of learning unified representations as a foundation for omnimodal modeling.1

This paper presents a data-driven approach to automatic blind equalization of audio by predicting log-mel spectral features and deriving an inverse filter. The method uses a deep neural network, where a pre-trained model provides semantic embeddings as a backbone, and only a lightweight head is trained. This design is intended to enhance training efficiency and generalization. Trained on both music and speech, the model is robust to noise and reverberation. Objective evaluations confirm its effectiveness, and subjective tests show performance comparable to that of an oracle that uses true log-mel spectral features, indicating that the model accurately estimates the desired characteristics, with remaining limitations attributed to the filtering stage. Overall, the results highlight the potential of the method for real-world audio enhancement applications.