arXiv_SD
-
Recurrence-Based Nonlinear Vocal Dynamics as Digital Biomarkers for Depression Detection from Conversational Speech
2026-04-29 02:51:35Himadri S SamantaAbstract
Digital biomarkers for depression have largely relied on static acoustic descriptors, pooled summary statistics, or conventional machine learning representations. Such approaches may miss nonlinear temporal organization embedded in conversational vocal dynamics. We hypothesized that depression is associated with altered recurrence structure in vocal state trajectories, reflecting changes in how the vocal system revisits acoustic states over time. Using the depression subset of the DAIC-WOZ corpus with 142 labeled participants, we modeled frame-level COVAREP trajectories as nonlinear dynamical systems and derived recurrence-based biomarkers from 74 vocal channels. Logistic regression with feature selection and stratified cross-validation evaluated classification performance. Recurrence-based biomarkers achieved a mean cross-validated AUC of 0.689, exceeding static acoustic baselines, entropy-dynamics features, Hurst exponent features, determinism features, and Lyapunov-like instability proxies. Permutation testing indicated statistical significance with $p=0.004$. Pooled cross-validated predictions yielded AUC 0.665 with a 95\% bootstrap confidence interval of [0.568, 0.758]. These findings suggest that depression may be characterized by altered recurrence structure in conversational vocal dynamics and support nonlinear state-space analysis as a promising direction for digital psychiatric biomarkers.
Abstract (translated)
URL
https://arxiv.org/abs/2604.26242
PDF
-
SymphonyGen: 3D Hierarchical Orchestral Generation with Controllable Harmony Skeleton
2026-04-28 11:01:21Xuzheng He, Nan Nan, Zhilin Wang, Ziyue Kang, Zhuoru Mo, Ao Li, Yu Pan, Xiaobing Li, Feng Yu, Xiaohong GuanAbstract
Generating symphonic music requires simultaneously managing high-level structural form and dense, multi-track orchestration. Existing symbolic models often struggle with a "complexity-control imbalance", in which scaling bottlenecks limit long-term granular steerability. We present SymphonyGen, a 3D hierarchical framework for contemporary cinematic orchestration. SymphonyGen employs a cascading decoder architecture that decomposes the Bar, Track, and Event axes, improving computational efficiency and scalability over conventional 1D or 2D models. We introduce "short-score" conditioning via a beat-quantized multi-voice harmony skeleton, enabling outline control while preserving textural diversity. The model is further refined using Group Relative Policy Optimization (GRPO) with a cross-modal audio-perceptual reward, aligning symbolic output with modern acoustic expectations. Additionally, we implement a dissonance-averse sampling algorithm to suppress unintended tonal clashes during inference. Objective evaluations show that both reinforcement learning and dissonance-averse sampling effectively enhance harmonic cleanliness while maintaining melodic expression. Subjective evaluations demonstrate that SymphonyGen outperforms baselines in musicality and preference for orchestral music generation. Demo page: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2604.25498
PDF
-
ML-SAN: Multi-Level Speaker-Adaptive Network for Emotion Recognition in Conversations
2026-04-28 08:51:29Kexue Wang, Yinfeng Yu, Liejun WangAbstract
To establish empathy with machines, it is essential to fully understand human emotional changes. However, research in multimodal emotion recognition often overlooks one problem: individual expressive traits vary significantly, which means that different people may express emotions differently. In our daily lives, we can see this. When communicating with different people, some express "happiness" through their facial expressions and words, while others may hide their happiness or express it through their actions. Both are expressions of 'happiness,' but such differences in emotional expression are still too difficult for machines to distinguish. Current emotion recognition remains at a 'static' level, using a single recognition model to identify all emotional styles. This "simplification" often affects the recognition results, especially in multi-turn dialogues. To address this problem, this paper introduces a novel Multi-Level Speaker Adaptive Network (ML-SAN), which, specifically, effectively addresses the challenge of speaker identity information confusion. ML-SAN does not simply assign a speaker's ID after recognition; instead, it employs a three-stage adaptive process: First, Input-level Calibration uses Feature-Level Linear Modulation (FiLM) to adjust the raw audio and visual features into a neutral space unrelated to the speaker. Then, Interaction-level Gating re-adjusts the trust level for each modality (e.g., voice or facial features) based on the speaker's identity information. Finally, Output-level Regularization maintains the consistency of speaker features in the latent space. Tests on the MELD and IEMOCAP datasets show that our model (ML-SAN) achieves better results, performs exceptionally well in handling challenging tail sentiment categories, and better addresses the diversity of speakers in real-world scenarios.
Abstract (translated)
URL
https://arxiv.org/abs/2604.25383
PDF
-
Hu'i S`u: Co-constructing a Dual Feedback Apparatus
Abstract
This performance presents a duet between two intelligent musical instruments, Sù (to trace back; to go upstream) and Agentier (playing on agentic clavier), and their human performers, connected through feedback loops. Rather than treating AI as a tool that responds predictably to input, both systems operate recursively, where past actions continuously influence future behaviour. The Sù operates in the audio space through latent representation. Its performer uses Make Noise 0-series synthesisers and MIDI controllers to work with a neural feedback synthesis system based on a RAVE model, with a latent feedback loop embedded within the model's internal structure. This allows the instrument to remember and reuse its own internal states, influencing ongoing sound generation through its recent sonic history. The Agentier functions in the control space. Its performer interacts with the system using a Roland S-1 synthesiser and Keith McMillen QuNeo touchpad, where control gestures are routed into a recurrent neural network that feeds back into the synthesis process. Through this feedback loop, the system actively shapes the evolution of control signals over time. Contrasting feedback in the audio and control domains, the performance explores shared agency, resistance, and negotiation between humans and intelligent musical systems. Musical phenomena are co-produced through the entangled states of interaction, rather than through pre-existing system configuration or fixed mappings.
Abstract (translated)
URL
https://arxiv.org/abs/2604.25207
PDF
-
S-SONDO: Self-Supervised Knowledge Distillation for General Audio Foundation Models
2026-04-27 19:20:47Mohammed Ali El Adlouni, Aurian Quelennec, Pierre Chouteau, Geoffroy Peeters, Slim EssidAbstract
General audio foundation models have recently achieved remarkable progress, enabling strong performance across diverse tasks. However, state-of-the-art models remain extremely large, often with hundreds of millions of parameters, leading to high inference costs and limited deployability on edge devices. Knowledge distillation is a proven strategy for model compression, but prior work in audio has mostly focused on supervised settings, relying on class logits, intermediate features, or architecture-specific techniques. Such assumptions exclude models that output only embeddings, such as self-supervised or metric-learning models. We introduce S-SONDO (Self-Supervised KnOwledge DistillatioN for General AuDio FOundation Models), the first framework to distill general audio models using only their output embeddings. By avoiding the need for logits or layer-level alignment, S-SONDO is architecture-agnostic and broadly applicable to embedding-based teachers. We demonstrate its effectiveness by distilling two audio foundation models into three efficient students that are up to 61 times smaller while retaining up to 96% of teacher performance. We also provide practical insights on loss choice and clustering-based balanced data sampling. Code is available here: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2604.24933
PDF
-
An event-based sequence modeling approach to recognizing non-triad chords with oversegmentation minimization
2026-04-27 12:17:13Leekyung Kim, Jonghun ParkAbstract
Automatic chord recognition (ACR) extracts time-aligned chord labels from music audio recordings. Despite recent advances, ACR still struggles with oversegmentation, data scarcity, and imbalance, especially in recognizing complex chords such as non-triads, which are unpopular in existing datasets. To address these challenges, we reformulate ACR as a segment-level sequence-to-sequence prediction task, where chord sequences are predicted auto-regressively rather than frame by frame. This design mitigates excessive segmentation by detecting chord changes only at segment boundaries. We further introduce two types of token representations and an encoder pre-training method, both specifically designed for time-aligned chord modeling. Experimental results show that our model improves performance in both chord recognition and segmentation, with notable gains for complex and infrequent chord types. These findings demonstrate the effectiveness of segment-level sequence modeling, structured tokenization, and representation learning for advancing chord recognition systems.
Abstract (translated)
URL
https://arxiv.org/abs/2604.24386
PDF
-
Keyword spotting using convolutional neural network for speech recognition in Hindi
2026-04-26 21:11:29Saru Bharti, Pushparaj Mani PathakAbstract
In this study, we investigate the application of keyword spotting (KWS) in the domain of Hindi speech recognition, utilizing a dataset comprising 40,000 audio samples. With a sampling rate of 44 kHz and an average duration of 1.9 seconds per sample, we focus on developing an efficient on-device KWS system tailored for user-specific queries. Leveraging Convolutional Neural Networks (CNNs) for classification, we employ feature engineering techniques to convert raw audio recordings into Mel Frequency Cepstral Coefficients (MFCCs) as an input for our network. Our experiments encompass various CNN architectures, exploring their efficacy in identifying predefined keywords within the continuous speech stream. Our CNN-based approach achieves a commendable accuracy rate of 91.79% through rigorous evaluation, demonstrating promising performance while ensuring computational efficiency and user-specific customization in Hindi speech recognition.
Abstract (translated)
URL
https://arxiv.org/abs/2605.02928
PDF
-
RTCFake: Speech Deepfake Detection in Real-Time Communication
2026-04-26 14:42:50Jun Xue, Zhuolin Yi, Yihuan Huang, Yanzhen Ren, Yujie Chen, Cunhang Fan, Zicheng Su, Yonghong Zhang, Bo CaiAbstract
With the rapid advancement of speech generation technologies, the threat posed by speech deepfakes in real-time communication (RTC) scenarios has intensified. However, existing detection studies mainly focus on offline simulations and struggle to cope with the complex distortions introduced during RTC transmission, including unknown speech enhancement processes (e.g., noise suppression) and codec compression. To address this challenge, we present the first large-scale speech deepfake dataset tailored for RTC scenarios, termed \textit{RTCFake}, totaling approximately 600 hours. The dataset is constructed by transmitting speech through multiple mainstream social media and conferencing platforms (e.g., Zoom), enabling precise pairing between offline and online speech. In addition, we propose a phoneme-guided consistency learning (PCL) strategy that enforces models to learn platform-invariant semantic structural representations. In this paper, the RTCFake dataset is divided into training, development, and evaluation sets. The evaluation set further includes both unseen RTC platforms and unseen complex noise conditions, thereby providing a more realistic and challenging evaluation benchmark for speech deepfake detection. Furthermore, the proposed PCL strategy achieves significant improvements in both cross-platform generalization and noise robustness, offering an effective and generalizable modeling paradigm. The \textit{RTCFake} dataset is provided in the {this https URL}.
Abstract (translated)
URL
https://arxiv.org/abs/2604.23742
PDF
-
Opening the Design Space: Two Years of Performance with Intelligent Musical Instruments
Abstract
Machine generation of symbolic music and digital audio are hot topics but there have been relatively few digital musical instruments that integrate generative AI. Present musical AI tools are not artist centred and do not support experimentation or integrating into musical instruments or practices. This work introduces an inexpensive generative AI instrument platform based on a single board computer that connects via MIDI to other musical devices. The platform uses artist-collected datasets with models trained on a regular computer. This paper asks what the design space of intelligent musical instruments might look like when accessible and portable AI systems are available for artistic exploration. I contribute five examples of instruments created and tested through a two-year first-person artistic research process. These show that (re)mapping can replace retraining for discovering AI interaction, that fast input interleaving is a new co-creative strategy, that small-data AI models can be a transportable design resource, and that cheap hardware can lower barriers to inclusion. This work could enable artists to explore new interaction and performance schemes with intelligent musical instruments.
Abstract (translated)
URL
https://arxiv.org/abs/2604.23583
PDF
-
An audio-to-analysis pipeline with certified transcription for information-theoretic profiling of the piano repertoire
Abstract
We present an audio-to-analysis pipeline that produces composer-level information-theoretic profiles : reflecting compositional vocabulary as it emerges from aggregated performances : from raw recordings, built on a transcription layer whose accuracy we certify on a standard benchmark (F1 = 0.9791 on the MAESTRO v3.0.0 test set). Applied to 1,238 pieces and 15 MAESTRO composers with at least ten attributed pieces, spanning the Baroque through the early twentieth century, the pipeline derives empirical distributions over harmonic scale degrees and analyzes them through Shannon entropy, asymmetric Kullback-Leibler divergence, and Zipfian rank-frequency modeling. The resulting profiles (i) order composers along an interpretable axis of harmonic predictability, with a narrow entropy range (3.33-3.86 bits) that reveals the marginal-level similarity of tonal vocabularies; (ii) recover known stylistic lineages (Haydn-Beethoven, Liszt-Rachmaninoff, Schubert-Schumann) through the smallest KL divergences in the corpus, with Mendelssohn emerging as a stable outlier within this corpus; and (iii) separate contemporary neoclassical artists (Richter, Frahm, Glass, Arnalds, Jóhannsson) from historical composers on the quality of Zipfian fit to the transition distribution, with mean $R^2 = 0.78$ for neoclassical versus 0.46 for historical (N $\geq$ 10 pieces each). This gap is larger than the spread within either group and is consistent with a minimalist compositional tendency: a compact transition vocabulary used with sharper frequency-rank regularity than historical composers. All estimates are reported with Laplace-smoothed bootstrap 95% confidence intervals.
Abstract (translated)
URL
https://arxiv.org/abs/2605.06685
PDF
-
Prompting Whisper for Joint Speech Transcription and Diarization
Abstract
As part of the MediSpeech project, we aim to develop a system that transcribes and diarizes Dutch conversations between doctors and patients in real-time. In this research (in-progress) we explore ways of efficiently combining Whisper with speaker diarization (SD). After trying to prompt Whisper with text that contains speaker labels, we observed that it is able to insert labels into the transcription with promising accuracy. We continued this line of research by fine-tuning Whisper with speaker-labelled prompts to generate transcriptions in a format similar to that of Serialized Output Training (SOT). Fine-tuning Whisper yielded more consistent speaker IDs across the chunks of long-form audio and improved verbatim transcription. The study uncovered new challenges as Whisper's SD performance suffers because of mistakes that get propagated through prompts and inaccurate timestamps assigned to overlapping speech.
Abstract (translated)
URL
https://arxiv.org/abs/2605.05231
PDF
-
Modulation Feature Enhancement with a Multi-Stage Attention Network for Underwater Acoustic Target Recognition
2026-04-24 18:10:57Shefeng Yan, Linlin Mao, Zeping Sui, Chunjin JiangAbstract
Underwater acoustic target recognition is critical for maritime applications, yet it faces challenges arising from the complex and diverse nature of ship-radiated noise. To address these issues, we propose a robust deep learning-based framework. First, we introduce a feature extraction and fusion method based on variational mode decomposition (VMD) and the 3/2-D spectrum to generate high-fidelity 2-D DEMON spectral features, which effectively capture modulation envelope information. To further enhance feature representation, we design a one-dimensional convolutional neural network (1-D CNN) integrated with a novel Multi-Stage Multi-Type Attention Mechanism (MMATT) that adaptively refines features at different network depths. Within this mechanism, we propose a Residual Channel-Independent Spectral Attention Mechanism (R-CISAM) and a Multi-Scale Separate-and-Fuse Spectral Attention Mechanism (MS-SFSAM). Moreover, to mitigate performance degradation caused by severe class imbalance inherent in real-world ship-radiated noise data, we devise an Adjustable Class-Balanced Focal Loss (ACBFL), which provides flexibility across tasks with varying degrees of imbalance. Experimental results on a real-world ship-radiated noise dataset demonstrate that the proposed solutions effectively enhance underwater acoustic target recognition performance.
Abstract (translated)
URL
https://arxiv.org/abs/2605.16304
PDF
-
Come Together: Analyzing Popular Songs Through Statistical Embeddings
2026-04-24 18:00:48Matthew Esmaili Mallory, Mark Glickman, Jason BrownAbstract
Statistical modeling of popular music presents a unique challenge due to the complexity of song structures, which cannot be easily analyzed using conventional statistical tools. However, recent advances in data science have shown that converting non-standard data objects into real vector-valued embeddings enables meaningful statistical analysis. In this work, we demonstrate an approach based on logistic principal component analysis to construct embeddings from global song features, allowing for standard multivariate analysis. We apply this method to a corpus of Lennon and McCartney songs from 1962-1966, using embeddings derived from chords, melodic notes, chord and pitch transitions, and melodic contours. Our analysis explores how these song embeddings cluster by Beatles album, how songwriting styles evolved over time, and whether Lennon and McCartney's compositions exhibited convergence or divergence. This embedding-based approach offers a powerful framework for statistically examining musical structure and stylistic development in popular music.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22925
PDF
-
Transformer-Based Rhythm Quantization of Performance MIDI Using Beat Annotations
2026-04-24 07:10:11Maximilian Wachter, Sebastian Murgul, Michael HeizmannAbstract
Rhythm transcription is a key subtask of notation-level Automatic Music Transcription (AMT). While deep learning models have been extensively used for detecting the metrical grid in audio and MIDI performances, beat-based rhythm quantization remains largely unexplored. In this work, we introduce a novel deep learning approach for quantizing MIDI performances using a priori beat information. Our method leverages the transformer architecture to effectively process synchronized score and performance data for training a quantization model. Key components of our approach include dataset preparation, a beat-based pre-quantization method to align performance and score times within a unified framework, and a MIDI tokenizer tailored for this task. We adapt a transformer model based on the T5 architecture to meet the specific requirements of rhythm quantization. The model is evaluated using a set of score-level metrics designed for objective assessment of quantization performance. Through systematic evaluation, we optimize both data representation and model architecture. Additionally, we apply performance and score augmentations, such as transposition, note deletion, and performance-side time jitter, to enhance the model's robustness. Finally, a qualitative analysis compares our model's quantization performance against state-of-the-art probabilistic and deep-learning models on various example pieces. Our model achieves an onset F1-score of 97.3% and a note value accuracy of 83.3% on the ASAP dataset. It generalizes well across time signatures, including those not seen during training, and produces readable score output. Fine-tuning on instrument-specific datasets further improves performance by capturing characteristic rhythmic and melodic patterns. This work contributes a robust and flexible framework for beat-based MIDI quantization using transformer models.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22290
PDF
-
Audio Effect Estimation with DNN-Based Prediction and Search Algorithm
2026-04-24 06:45:35Youichi Okita, Haruhiro KatayoseAbstract
Audio effects play an essential role in sound design. This research addresses the task of audio effect estimation, which aims to estimate the configuration of applied effects from a wet signal. Existing approaches to this problem can be categorized into predictive approaches, which use models pre-trained in a data-driven manner, and search-based approaches, which are based on wet signal reconstruction. In this study, we propose a novel approach that integrates these approaches: first, DNNs predict the dry signal and effect configuration, and then a search is performed based on wet signal reconstruction using these predictions. By estimating the dry signal in the prediction stage, it becomes possible to complement or improve the predictions using reconstruction similarity as an objective function. The experimental evaluation showed that methods based on the proposed approach outperformed the method solely based on the predictive approach. Furthermore, the findings suggest that the task division of predicting the effect type combination followed by the search-based estimation of order and parameters was the most effective across various metrics.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22276
PDF
-
Advancing automatic speech recognition using feature fusion with self-supervised learning features: A case study on Fearless Steps Apollo corpus
2026-04-24 04:15:43Szu-Jui Chen, John H. L. HansenAbstract
Using self-supervised learning (SSL) models has significantly improved performance for downstream speech tasks, surpassing the capabilities of traditional hand-crafted features. This study investigates the amalgamation of SSL models, with the aim to leverage both their individual strengths and refine extracted features to achieve improved speech recognition models for naturalistic scenarios. Our research investigates the massive naturalistic Fearless Steps (FS) APOLLO resource, with particular focus on the FS Challenge (FSC) Phase-4 corpus, providing the inaugural analysis of this dataset. Additionally, we incorporate the CHiME-6 dataset to evaluate performance across diverse naturalistic speech scenarios. While exploring previously proposed Feature Refinement Loss and fusion methods, we found these methods to be less effective on the FSC Phase-4 corpus. To address this, we introduce a novel deep cross-attention (DCA) fusion method, designed to elevate performance, especially for the FSC Phase-4 corpus. Our objective is to foster creation of superior FS APOLLO community resources, catering to the diverse needs of researchers across various disciplines. The proposed solution achieves an absolute +1.1% improvement in WER, providing effective meta-data creation for the massive FS APOLLO community resource.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22203
PDF
-
Beyond Acoustic Sparsity and Linguistic Bias: A Prompt-Free Paradigm for Mispronunciation Detection and Diagnosis
2026-04-24 00:38:00Haopeng Geng, Longfei Yang, Xi Chen, Haitong Sun, Daisuke Saito, Nobuaki MinematsuAbstract
Mispronunciation Detection and Diagnosis (MDD) requires modeling fine-grained acoustic deviations. However, current ASR-derived MDD systems often face inherent limitations. In particular, CTC-based models favor sequence-level alignments that neglect transient mispronunciation cues, while explicit canonical priors bias predictions toward intended targets. To address these bottlenecks, we propose a prompt-free framework decoupling acoustic fidelity from canonical guidance. First, we introduce CROTTC, an acoustic model enforcing monotonic, frame-level alignment to accurately capture pronunciation deviations. Second, we implicitly inject mispronunciation information via the IF strategy under the knowledge transfer principle. Experiments show CROTTC-IF achieves a 71.77% F1-score on L2-ARCTIC and 71.70% F1-score on the Iqra'Eval2 leaderboard. With empirical analysis, we demonstrate that decoupling acoustics from explicit priors provides highly robust MDD.
Abstract (translated)
URL
https://arxiv.org/abs/2604.22133
PDF
-
Spectrographic Portamento Gradient Analysis: A Quantitative Method for Historical Cello Recordings with Application to Beethoven's Piano and Cello Sonatas, 1930--2012
Abstract
Portamento in string performance has been studied primarily as a binary presence-or-absence phenomenon, with existing research measuring frequency of occurrence and, less commonly, duration in milliseconds. This paper introduces a third quantitative descriptor; the spectrographic gradient of the portamento slide, measured in Hz/second, and demonstrates its measurement using a protocol combining Sonic Visualizer's melodic spectrogram layer, GIMP pixel analysis, and metric calibration against the spectrogram's known frequency axis. The gradient captures what duration alone cannot: the steepness of the pitch trajectory, which encodes the expressive character of the slide independently of its length. Applied to the opening measures of. Specifically because their monophonic texture permits reliable spectrographic pitch tracking. The method yields gradient values ranging from approximately 600~Hz/s in late-period recordings to over 4,000~Hz/s in early twentieth-century performances. The paper further documents a gain-recovery protocol that extends the analysable corpus to analogue recordings from the 1930s where portamento traces are faint in digital transfer. Applying the method to a corpus of 22 recordings spanning 1930--2012, the paper tests the hypothesis that gradient steepness correlates negatively with tempo: that slower performances produce steeper, longer slides while faster performances produce shallower slides or none at all. The results support this hypothesis, suggesting that the widely documented decline of portamento across the twentieth century is not a binary transition from presence to absence but a continuou
Abstract (translated)
URL
https://arxiv.org/abs/2604.22037
PDF
-
Beyond Rules: Towards Basso Continuo Personal Style Identification
Abstract
A central part of the contemporary Historically Informed Practice movement is basso continuo, an improvised accompaniment genre with its traditions originating in the baroque era and actively practiced by many keyboard players nowadays. Although computational musicology has studied the theoretical foundations of basso continuo expressed by harmonic and voice-leading rules and constraints, characteristics of basso continuo as an active performing art have been largely overlooked mostly due to a lack of suitable performance data that could be empirically analyzed. This has changed with the introduction of The Aligned Continuo Realization Dataset (ACoRD) and the basso continuo realization-to-score alignment. Basso continuo playing is shaped by stylistic traditions coming from historical treatises, but it also may provide space for showcasing individual performance styles of its practitioners. In this paper, we attempt to explore the question of the presence of personal styles in the basso continuo realizations of players in the ACoRD dataset. We use a historically informed structured representation of basso continuo performance pitch content called griffs and Support Vector Machines to see whether it is possible to classify players based on their performances. The results show that we can identify players from their performances. In addition to the player classification problem, we discuss the elements that make up the individual styles of the players.
Abstract (translated)
URL
https://arxiv.org/abs/2604.21822
PDF
-
Time vs. Layer: Locating Predictive Cues for Dysarthric Speech Descriptors in wav2vec 2.0
2026-04-23 12:46:49Natalie Engert, Dominik Wagner, Korbinian Riedhammer, Tobias BockletAbstract
Wav2vec 2.0 (W2V2) has shown strong performance in pathological speech analysis by effectively capturing the characteristics of atypical speech. Despite its success, it remains unclear which components of its learned representations are most informative for specific downstream tasks. In this study, we address this question by investigating the regression of dysarthric speech descriptors using annotations from the Speech Accessibility Project dataset. We focus on five descriptors, each addressing a different aspect of speech or voice production: intelligibility, imprecise consonants, inappropriate silences, harsh voice and monoloudness. Speech representations are derived from a W2V2-based feature extractor, and we systematically compare layer-wise and time-wise aggregation strategies using attentive statistics pooling. Our results show that intelligibility is best captured through layer-wise representations, whereas imprecise consonants, harsh voice and monoloudness benefit from time-wise modeling. For inappropriate silences, no clear advantage could be observed for either approach.
Abstract (translated)
URL
https://arxiv.org/abs/2604.21628
PDF
