arXiv_SD
-
Automatic Transcription of Microtonal Free-Rhythm Vocal Music: A Case Study in Iranian Classical Music
2026-08-17 20:41:34Sepideh Shafiei, Shapour Hakam, Harsh Dange, Joel Rodriguez CaraballoAbstract
This paper introduces a computational workflow for automatically transcribing microtonal, free-rhythm vocal music, with Iranian classical music as a case study. Our approach is based on performances by the renowned vocalist Karimi and ground truth transcriptions by the prominent ethnomusicologist Masoudieh [14], which were subsequently incorporated into the IRMA Audio-MIDI dataset [20]. To accurately extract melodies, we employ pitch histograms in conjunction with Dynamic Time Warping (DTW). Additionally, we introduce specialized musical notations to capture the intricate ornamentations characteristic of the genre, with particular emphasis on the vocal technique tahrir. The transcription process is implemented in Python using the music21 library for symbolic music representation [5]. This study not only advances the field of computational ethnomusicology but also highlights the potential of computational methods in preserving and analyzing complex musical traditions. The transcription system also generates a combined visualization of the audio pitch contour and the DTW-aligned MIDI representation, enabling users to inspect the correspondence between the performance and the generated transcription. A companion visual editor supports expert-in-the-loop correction of the resulting notation.
Abstract (translated)
URL
https://arxiv.org/abs/2608.17114
PDF
-
A Multiplication-Free Feature Extractor for Signal Classification: Keyword Spotting Case Study
Abstract
A very low complexity feature extractor called next iRDT is proposed and evaluated for the problem of keyword spotting (KWS). Unlike any other types of feature extractors including the widely used MFCC, or adaptive, CNN-based ones, our algorithm is multiplier-free and it employs only simple, energy-efficient arithmetic operators. Since keyword-spotting of speech commands (KWS) is a typical application for TinyML platforms requiring low complexity for the signal classification chain, we consider it as a case study to evaluate complexity and functional performance. If properly tuned, iRDT demonstrates similar accuracy to solutions based on MFCC or CNN-based extractors using baseline classifiers on Google's KWS 12-classes dataset. With a different classifier the system achieved 94.7% validation accuracy. Processing times on CPU for the proposed feature extractor, are at least one order of magnitude smaller than for the MFCC. The proposed algorithm has a very low hardware footprint, making it ideal for ultra-low power edge devices. Code and demo are available [18].
Abstract (translated)
URL
https://arxiv.org/abs/2608.17108
PDF
-
How Fragile Is Your Watermark? Training-Free Structural Removal of Neural Audio Watermarks
Abstract
Neural audio watermarks are increasingly used to attribute and detect AI-generated speech, so their practical value rests on how cheaply an adversary can remove them. Robustness is usually measured by running a fixed battery of distortions blindly against every scheme. We instead make removal diagnostic: from a few clean/watermarked pairs we compute cheap structural probes that reveal where a watermark sits in the signal (its embedding domain), then apply a single domain-matched attack rather than a blind sweep. We further summarize each scheme with one threshold-free fragility score, the area under its accuracy-versus-quality trade-off, which an accuracy-only benchmark cannot provide. Across ten watermarking schemes the probes separate fragile from robust marks: for magnitude and carrier-domain watermarks a single matched attack erases the payload (WavMark, SilentCipher, audiowmark) or removes the detection flag (AudioSeal) at high objective quality (PESQ >= 3.6), whereas latent-domain marks (VoiceMark, WMCodec, AlignMark, AWARE) resist every training-free attack we apply. The same pair-only probe signatures also identify which watermarking scheme is present (84% over ten schemes).
Abstract (translated)
URL
https://arxiv.org/abs/2608.16566
PDF
-
Sonifying I2S Transport Signals to Detect Transmission Faults
Abstract
This paper outlines a sonification design to support fault detection in the transmission of I2S transport signals. I2S is a protocol for communicating real-time digital audio between integrated circuits that, while in wide and general use, does not include built-in error detection. Moreover, given the nature of the protocol transmission faults affecting timing, framing and alignment can be difficult to identify using conventional visual methods. The proposed design addresses this with an approach informed by Audification, wherein oversampling controls temporal rescaling to render protocol structure (SCK and WS) and payload data (SD) across separate stereo channels. A preliminary computational feasibility study was carried out to measure feature-space separability of I2S faults in the generated auditory representations as opposed to listener performance. It evaluates the design across several payload types and error conditions including jitter, bit-slip, and word-length errors. Class separability was assessed through clustering analyses of extracted features. The evaluation results show that while oversampling produces systematic changes in feature values, it does not meaningfully improve separability between error classes. However, a modest but consistent improvement in separability is observed as a function of the joint representation of structural and payload information across channels. The findings suggest that feature-space separability in sonified communication protocol data may be dependent on the integration of complementary information streams, rather than on signal scaling alone.
Abstract (translated)
URL
https://arxiv.org/abs/2608.16498
PDF
-
Contrastive Learning with Variational Regularization for Multi-Session EEG-to-Speech Decoding
2026-08-17 10:11:11Tomoaki Mizuno, Toru NakashikaAbstract
Reconstructing heard speech from non-invasive electroencephalography (EEG) is challenging due to a low signal-to-noise ratio (SNR) and inter-session variability. While trial averaging improves the SNR, it is difficult to apply to continuous speech. We instead use repeated EEG responses to the same stimulus across different sessions as positive pairs for contrastive learning, and introduce variational regularization that, combined with this contrastive objective, keeps the encoder representation space broad. Experiments on a Japanese EEG dataset show that combining the session-invariant strategy with variational regularization improves the character error rate (CER) while maintaining mel-spectrogram reconstruction fidelity. Session probing confirms that the encoder representations achieve session-invariance.
Abstract (translated)
URL
https://arxiv.org/abs/2608.16360
PDF
-
A Novel Binaural Cue Preservation Loss for DNN-Based Binaural Speech Enhancement
2026-08-17 09:12:44Jayteerth Amble, Thomas Haubner, Hendrik Schr\"oter, Christoph Hoog Antink, Henning PuderAbstract
Binaural speech enhancement for hearing aids aims to reduce noise while preserving the interaural cues needed for spatial localization. Although deep neural network-based methods achieve strong noise reduction, they often distort the rela- tionship between the left and right signals. In this paper, we propose two novel binaural cue preservation losses. First, a binaural reconstruction error loss that directly penalizes masking-induced distortion in the relationship between the left and right spectra, providing a more direct measure of the binaural consistency than conventional separate interaural level differences (ILD) and interaural phase differences (IPD) errors as in prior work. Second, a binaural cue loss that jointly models ILD and IPD to better preserve the binaural structure. Experimental results show that both proposed losses maintain strong noise reduction performance and reduce masking- induced distortion compared to the state-of-the-art baseline cue loss, while the second proposed joint binaural cue loss also outperforms the baseline in ILD preservation.
Abstract (translated)
URL
https://arxiv.org/abs/2608.16299
PDF
-
Geometry-adaptive Ambisonic encoding for sparse microphone arrays of variable topology using physics-informed diffusion
2026-08-17 08:17:21Xiang Zhou, Zhengqiao Zhao, Zhengding Luo, Wen ZhangAbstract
Ambisonics delivers compact scene based spatial audio representation, yet higher order Ambisonic encoding poses difficulties for wearables and embedded hardware. Their microphone arrays are often sparse, irregular, and constrained by device specific boundary conditions. These factors make the spherical-harmonic (SH) domain encoding ill conditioned: inverse filtering amplifies noise, while deterministic neural encoders may overfit to array-specific responses or smooth ambiguous higher-order components. This paper presents DiffM2A, a geometry-adaptive conditional diffusion framework for robust Ambisonic encoding from sparse MAs with variable topologies. Its Geometry-Adaptive Spherical Harmonic Projection (GASHP) front-end constructs boundary-aware SH steering functions and applies an energy-normalized modal projection, mapping array-dependent observations to a common modal representation without explicit pseudo-inverse computation. A dual-branch Elucidated Diffusion Model then estimates complex Ambisonic coefficients, conditioned on both the raw microphone spectra and GASHP features. Sound intensity and rotational equivariance losses further enhance inter-channel phase consistency and structured behavior across SH subspaces. Evaluations on both first- and second-order Ambisonic encoding tasks, using simulated room-acoustics and real-world LOCATA recordings, demonstrate that DiffM2A outperforms conventional and neural baseline methods on signal fidelity, spectral accuracy, spatial coherence, and binaural cue preservation. Additional experiments show that these gains are largely retained across unseen five-microphone layouts and under mismatched open-array and rigid-sphere boundary models.
Abstract (translated)
URL
https://arxiv.org/abs/2608.16240
PDF
-
ACE-Cap: Active Evidence Acquisition via Agentic Co-Evolution for Long-Paragraph Fine-Grained Audio Captioning
2026-08-17 06:26:17Fengji Ma, Yan Rong, Xu Li, Xuenan Xu, Chen Zhang, Li LiuAbstract
Long-paragraph fine-grained audio captioning requires models to recover diverse acoustic facts while avoiding omissions and unsupported details. However, prevailing captioners remain passive one-shot generators: once a detail is overlooked, they cannot identify the evidence gap, query the audio for targeted information, or decide when sufficient evidence has been collected. We formulate this task as active evidence acquisition and introduce Agentic Co-Evolution for Captioning (ACE-Cap). The framework uses multi-turn interaction between a Composer and an Instruct model to form a closed evidence-acquisition loop. A Captioner first produces an initial description. Conditioned on this description and the interaction history, a text-only Composer asks targeted questions about unresolved acoustic attributes, while an audio-conditioned Instruct model provides grounded answers. The Composer then decides when to terminate and synthesizes the accumulated evidence into a final caption. ACE-Cap trains these roles through a unified gold-to-prediction reward derived from fixed, gold-grounded multiple-choice questions and a frozen caption-only judge. For credit assignment in variable-length interactions, LOOP-GRPO replaces the trajectory-wide scalar advantage with span-aligned signals: leave-one-out contributions of individual questions to the accumulated evidence, a quality-cost utility for stopping, and an evidence-preservation utility for final synthesis. Role-wise warm-up followed by alternating Composer and Instruct optimization keeps each update a well-defined single-policy problem while allowing the roles to co-evolve. ACE-Cap thus turns captioning from passive one-shot generation into an adaptive process that learns what evidence to acquire, when to stop, and how to preserve it in a long-paragraph caption.
Abstract (translated)
URL
https://arxiv.org/abs/2608.16162
PDF
-
CineDub: Scaling End-to-End Video Dubbing to Multi-Speaker Dialogues with Coherent Sound Effects
2026-08-16 13:16:15Yusheng Dai, Kangdi Wang, Baolong Gao, Yuxuan Jiang, Weiqiang Wang, Qiuhong Ke, Jianfei CaiAbstract
Automatic video dubbing in the wild remains fundamentally limited by two competing constraints: hierarchical methods depend on brittle, multi-stage preprocessing pipelines that severely restrict data scalability and practical deployment, while holistic approaches operating on uncropped video suffer from weak temporal alignment and speaker-utterance ambiguity in multi-speaker settings. To overcome these limitations, we propose CineDub, a unified diffusion-based model that achieves precise multi-speaker dialogue dubbing directly from uncropped videos, without face cropping or speaker diarization. Central to our approach is the Implicitly-Coupled Holistic Conditioning (ICHC) paradigm, where holistic visual representations and a semantic-bundled transcription format are encoded independently, yet implicitly coupled through cross-modal training to resolve speaker ambiguity and enable precise multi-speaker multi-turn dialogue dubbing. Building on the unified temporal cues captured by holistic visual features, we further extend CineDub to joint speech and audio generation. We introduce an Ambient-to-Linguistic Curriculum Learning (ALC) to mitigate sub-task degradation, and a decoupled textual branch control mechanism to resolve cross-prompt interference during simultaneous generation. We also release two in-the-wild benchmarks, CineDub-Multi for multi-speaker dialogue dubbing and CineDub-SA for video-to-speech-and-audio (V2SA) generation, to enable evaluation under realistic conditions. Experiments show that CineDub achieves state-of-the-art results on established single-speaker dubbing and video-to-audio benchmarks while excelling in multi-speaker dialogue dubbing and acoustically coherent joint generation.
Abstract (translated)
URL
https://arxiv.org/abs/2608.15734
PDF
-
'What's That Sound?': A Versatile, Robust, and Lightweight Convolutional Transformer for Environment Sound Recognition
Abstract
The conventional hearing aid is both costly and lim- ited in usage, as it is not intended to detect non-speech audio. Our objective is to develop a machine learning solution to provide a more accurate and affordable mechanism to identify surrounding sounds to improve the safety of the hearing impaired, i.e., if a car is honking behind pedestrians, or a gunshot is fired, and they need to move away from the source. By adding randomized augmentations to audio, concatenating a Mel-Frequency Cepstral Coefficients (MFCCs) diagram and a log-mel Spectrogram, and including Convolutional Neural Networks (CNNs) in a Trans- former architecture, the Randomized Audiomentational Layered Convolutional Transformers (RALCT) model efficiently extracts features from diversified audio representations. In addition, RALCT is small enough, with only approximately 310,000 parameters, to be deployed into mobile devices. Experimental results on the UrbanSound8K dataset resulted in an accuracy consistently over 93% for all variations of RALCT with the highest at 94.56%, reaching state-of-the-art levels. To leverage the capabilities of this technology, a mobile app is developed to be integrated with the model to provide real-time safety control. RALCT thus represents a robust, lightweight, affordable, and versatile deep learning tool to aid the navigation and safety of the hearing impaired.
Abstract (translated)
URL
https://arxiv.org/abs/2609.27762
PDF
-
Prototype-Rectified Iterative Self-supervised Manifold Denoising under Severe Acoustic Shift
2026-08-15 04:42:14Ashish Anand Shukla, Rini Smita Thakur, Aryan Das, Vinod K. KurmiAbstract
Audio-Text Foundation Models (ATMs) fail catastrophically under severe acoustic noise, yet existing adaptation strategies either rely on gradient-based Test-Time Adaptation (TTA), which reinforces noise rather than signal, or on prompt tuning that requires privileged noise annotations unavailable at inference. We address these failures with PRISM (Prototype-Rectified Iterative Self-supervised Manifold Denoising), a training-free, source-free TTA framework grounded in the Affine Noise Hypothesis: severe acoustic noise induces a low-rank affine shift in the multimodal latent space, with more than 90% of distortion energy confined to the leading 60 principal components. PRISM estimates and reverses this distortion from an unlabeled target batch using frozen text prototypes as geometric anchors via three closed-form geometric corrections compiled into a single static projection matrix by Affine Bias Regression. At inference, adaptation reduces to one matrix-vector multiplication in 0.0009 ms, making it substantially faster than gradient-based TTA while requiring no additional training. On UrbanSound8K, PRISM improves over the zero-shot baseline by 12.94 percentage points and surpasses an oracle-assisted TTA baseline by 9.41 percentage points, despite never observing its privileged augmented noise prompts. We further identify the Polyphonic Trap, a principled failure mode of subspace deflation for broadband classes, and resolve it via Confidence-Aware Regression (CAR), recovering up to 8.16 percentage points for the worst-affected class.
Abstract (translated)
URL
https://arxiv.org/abs/2608.15037
PDF
-
A Parameter-Free Few-Shot Evaluation for Elephant Vocalisation Classification
2026-08-14 18:54:24Christiaan M. Geldenhuys, Thomas R. NieslerAbstract
We present a parameter-free episodic evaluation of nearest-centroid classification for elephant vocalisations on fixed pretrained acoustic embeddings, across the Elephant Voices (EV) and Linguistic Data Consortium (LDC) datasets. Rather than asking which embedding yields the best classifier when trained on all available labelled data, we ask how the simplest classifier performs as labelled exemplars per class are varied. Each class is represented by the mean of its support-set embeddings, and each query is assigned to the nearest centroid under squared Euclidean distance. We evaluate this centroid classifier on the Perch (ver. 1), Perch (ver. 2), and HuBERT (base, layer 2) embeddings, together with mel frequency cepstral coefficient (MFCC) features, in an N-way k-shot manner under the same cross-validation protocol as the trained baselines. A bootstrap over 100 resampled support sets quantifies the sampling noise. On the smaller, low-resource EV dataset, the centroid classifier using the stronger Perch (ver. 1) and Perch (ver. 2) embeddings overtakes the fully-trained logistic regression classifier from a single exemplar per class and the stronger recurrent classifier from two. Over the reduced set of call types on which the strongly-supervised end-to-end baseline was trained, the centroid classifier matches and then surpasses that baseline in mean average precision (mAP), from a few exemplars per class. On the larger LDC dataset, where labelled exemplars are abundant, the trained baselines retain their advantage at every k considered. At five exemplars per class, the centroid classifier using the strongest embedding, Perch (ver. 2), attains a mAP of 0.542 on the EV dataset and 0.368 on the LDC dataset. Parameter-free nearest-centroid classification is the stronger choice when labelled exemplars are few and the fixed embedding already encodes the features that separate the call types.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14824
PDF
-
What Makes a Good Layer? Assessing the Layer-Wise Intrinsic Properties of Music Foundation Models
2026-08-14 18:46:22Angelos-Nikolaos Kanatas, Yuexuan Kong, Pablo Alonso-Jim\'enez, Xavier Serra, Dmitry BogdanovAbstract
Music foundation models are commonly used as frozen audio feature extractors, yet selecting which layer to extract from remains largely heuristic. Current practice defaults to fixed depths or multi-layer fusion, with limited understanding of why certain layers transfer better across downstream tasks or how representation quality varies with depth and pre-training paradigm. We conduct a systematic layer-wise analysis of 12 music foundation models spanning three pre-training paradigms (masked modeling, autoregressive modeling, and contrastive learning), characterizing their hidden representations through intrinsic geometric and transformation-based properties. Correlating label-free representation-quality metrics with layer-wise performance across 15 downstream tasks, we find that several metrics track layer quality for genre classification, emotion recognition, automatic tagging, and beat tracking, albeit with varying strength across tasks and pre-training paradigms. However, all metrics fail on tonal tasks such as key estimation and chord recognition, indicating that no single property serves as a general proxy for representation quality across music information retrieval tasks. To address this gap, we introduce a pitch-transposition equivariance measure that captures properties missed by these standard metrics, providing a consistent indicator of tonal quality across model families. Finally, we show that intrinsic metrics can serve as effective proxies for layer selection, matching or outperforming trainable multi-layer fusion methods, particularly in limited-data settings.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14819
PDF
-
Singer-Informed Vocal Source Separation for Multi-Singer Music Mixtures
Abstract
Music source separation systems typically extract a single vocal track and do not distinguish between multiple singers. We study singer-informed vocal source separation for multi-singer mixtures. Our framework introduces a short enrollment recording of a target singer to guide separation through a learned embedding. The singer embedding is incorporated using feature concatenation or feature-wise linear modulation (FiLM), enabling the model to focus on the target singer while suppressing interference. We construct a duet dataset based on DAMP-VSEP with quality filtering and non-overlapping enrollment segments. Experiments on solo and duet settings show that while baseline models perform well for single-singer mixtures, the proposed method improves target-singer extraction in multi-singer cases, increasing target-singer SI-SDR from 0.33 dB to 5.58 dB. Fréchet Audio Distance (FAD) further shows improved perceptual quality and better alignment with target audio distributions. Code and checkpoints are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14516
PDF
-
AT-ADD: All-Type Audio Deepfake Detection Challenge Summary
2026-08-14 12:28:54Yuankun Xie, Haonan Cheng, Jiayi Zhou, Xiaoxuan Guo, Tao Wang, Changhao Zhang, Jian Liu, Weiqiang Wang, Ruibo Fu, Xiaopeng Wang, Hengyan Huang, Xiaoying Huang, Long Ye, Guangtao ZhaiAbstract
This paper summarizes the ACM Multimedia 2026 AT-ADD Grand Challenge on all-type audio deepfake detection. AT-ADD contains two tracks: robust speech deepfake detection under realistic acoustic and channel variations, and type-agnostic detection over speech, environmental sound, singing voice, and music. We describe the challenge tasks, dataset and evaluation-set design, official leaderboard results, and common design patterns observed in participating systems. The best Track 1 system achieved 90.71% Macro-F1 on the final evaluation set, while the best Track 2 system achieved 96.10% Macro-F1. The final submissions show that strong systems commonly combine large-scale self-supervised audio representations, data augmentation, multi-crop inference, and structured fusion or routing. The results also reveal remaining challenges in generalization to unseen generators, robustness to realistic speech-domain distortions, and balanced performance across heterogeneous audio types.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14249
PDF
-
RDSE: Real-Time Directional Speech Enhancement Framework for Dual-Mic Commodity Smartphones
2026-08-14 09:03:34Chinmay Deo, Danish Mahajan, Soudeep Ghoshal, Himanshu BuckchashAbstract
Audio-visual zooming on mobile devices is fundamentally constrained by limited microphone count, compute resource, unknown array geometry, and weak assumptions on real-world acoustic conditions. We present RDSE: Real-Time Directional Speech Enhancement Framework, a hybrid pipeline that combines traditional signal processing and modern machine learning methods for dual-microphone low-end smartphones. Unlike tightly coupled AV systems that continuously use visual feeds or embeddings, RDSE decouples visual interaction from feature inference, using camera input solely to parameterize the steering vector for spatial filtering. The proposed system uses four modules: dereverberation network, a beamformer with a temporal convolutional network, and a MetricGAN based denoiser. Extensive evaluation on benchmark data and real-world recordings demonstrates effective zooming without precise geometry calibration. Comparative experiments show competitive performance to existing methods. Proposed system is rigorously tested for statistical validity, ablation study, latency profiling, and practicality through real-time live-streaming.
Abstract (translated)
URL
https://arxiv.org/abs/2610.12476
PDF
-
Ambisonics Encoding of Room Impulse Responses using a Device-Agnostic Diffusion Mode
2026-08-14 08:57:32Eloi Moliner, Christoph Hold, Juan Azcarreta Ortiz, Sebastian Prepelita, Ishwarya Ananthabhotla, Daniel Wong, Sanjeel Parekh, Sanha LeeAbstract
We address the problem of encoding room impulse responses (RIRs) into high-order Ambisonics (HOA) representations from arbitrary and potentially insufficient or incomplete microphone array measurements. This task is fundamentally ill-posed for microphone arrays with limited spatial capture capabilities, such as irregular or sparse arrays, as classical linear methods fail to reconstruct high-order spatial detail. We introduce a diffusion-based generative framework that models the statistical properties of HOA RIRs. This enables device-agnostic encoding from arbitrary microphone arrays, potentially unseen during data measurement. Our approach incorporates a posterior sampling procedure that enforces consistency between the estimated signals and the measurements while plausibly reconstructing spatial information that is unobservable from the limited measurements alone. Experiments on simulated data demonstrate that our method outperforms linear and neural baselines, achieving accurate HOA RIR estimation up to 12th order. A listening test with binaural renderings, including both simulated and measured RIRs, further confirms that the proposed method yields higher perceptual similarity to reference Ambisonics RIRs than all baselines. The flexibility and accuracy of the proposed framework opens new possibilities for scalable acoustics simulations.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14097
PDF
-
The Note-Chord-Voice Framework: Structured Source Separation and Causal Inference for EV Charging Data
Abstract
Real-world EV charging data exhibit three interlocking pathologies: hardware fragmentation (network timeouts and billing resets split sessions), physical violations (independent energy/duration models produce impossible states like 50 kWh in 10 min on a 7 kW charger), and collider bias (clustering on post-treatment outcomes opens backdoor paths for price elasticity). We propose the Note-Chord-Voice framework, a music-inspired, axiom-driven pipeline that separates data cleaning (Repair Chords), structural pattern discovery (Harmonic Chords), descriptive source separation (NMF Voices), and causal inference into distinct, falsifiable stages. Key innovations: (i) falsification gates (A1-A5, G3, G10) that test data suitability before modeling; (ii) Gamma-initialized NMF with input rescaling for convergence stability from STL decomposition; (iii) tag-based coupon grading (A/B/C/D) to isolate quasi-random treatment from night-time confounders and targeted promotions; (iv) separate per-voice OLS to avoid simplex collinearity; (v) Foote novelty curves for structural regime detection. Applied to the Jiangmen dataset (495,707 sessions, 20 stations, from July 2024 to March 2025), all core axioms pass except G3 (no strong 168 h cycle). NMF achieves R^2=0.9921; the physically constrained duration model yields aggregate R^2=0.5409. Two voices are price-sensitive (beta = -11 to -14 min, p<0.001), of which one is stable (Voice 3, beta=-14.16) and one treatment-driven (Voice 1, beta=-11.10); only the stable voice supports causal claims. Counterfactual simulation shows targeting discounts to price-sensitive voices recovers 52.8% of discount expenditures (~0.85M CNY/year); restricting to the single stable price-sensitive voice yields a more conservative estimate.
Abstract (translated)
URL
https://arxiv.org/abs/2608.14756
PDF
-
Common Cause, Not Cross-Attention: Blocking Visual Shortcuts in Audio-Video Generation
2026-08-14 06:59:55Jian Xu, Delu Zeng, John Paisley, Qibin ZhaoAbstract
Joint audio--video generators are trained on data in which what an event looks like and what it sounds like are strongly, often spuriously, correlated: a particular material, texture, or object appearance co-occurs with a particular sound. This paper is a controlled causal study of the resulting failure mode. Building an AV structural causal model in which the audio is, by construction, independent of the video's nuisance appearance, we show that models which let audio read video directly-through cross-attention or a shared latent-learn a visual shortcut: they predict sound from appearance rather than from the causal event, and collapse when the appearance-event correlation is broken at test time, literally synthesizing the wrong event's sound. Crucially, the popular remedy of routing both modalities through a shared common-cause latent does not fix this: a bottleneck, an unsupervised shared/private factorization, and a faithful shared-prior model all grab the appearance proxy and fail like the direct model. Blocking the shortcut instead requires an intervention on the nuisance. Under the stated SCM and intervention assumptions we prove that counterfactual invariance is necessary and sufficient to identify the causal predictor, and we verify the mechanism across a feature-vector SCM, procedural pixel video, real images with spectrogram audio and a pretrained backbone, moving real digits, and a conditional generator. On a \emph{real, pretrained} video-to-audio generator, an input-intervention test shows the model is far from invariant to sound-irrelevant edits (recolouring or graying a video substantially changes the sound it generates).
Abstract (translated)
URL
https://arxiv.org/abs/2609.22361
PDF
-
H2H Music Improv: A Communication Model and Audio-Visual Dataset for Music Improvisation
2026-08-14 04:56:25Aleksandra Teng Ma, Anthony Cammarota, Jiayi Wang, Alexandria Smith, Cheng-Zhi Anna Huang, Jeffrey Albert, Alexander LerchAbstract
Current real-time AI improvisation systems lack the communication awareness human musicians rely on: rather than treating communication as a foundational algorithm design concern, most systems layer interaction strategies post-hoc onto generative algorithms through explicit controls and predefined modes. This gap persists in part because no formalized, machine-readable communication model with musicians exists. To address this, we study how expert musicians communicate in free (non-idiomatic) improvisation, unconstrained by prior discussion or agreement. Through a collaborative co-design process with expert improvisers, we derive a communication model that (1) captures how free improvisers negotiate musical ideas and enter stable musical spaces, and (2) is formalized as a machine-readable annotation scheme. We further present the H2H (Human-to-Human) Music Improvisation dataset: six hours of audio-visual expert duo improvisations with clean per-player stems and per-player annotations of both their own intentions and their perception of their partner's intentions. To our knowledge, this is the first such dataset for free improvisation. Together, the communication model and the dataset offer a new lens and resource for studying musician communication and may in future inform the design of AI musical partners that communicate by design.
Abstract (translated)
URL
https://arxiv.org/abs/2608.13957
PDF
