arXiv_SD
-
Listening Like a Judge: A Music-Aware Framework for Automatic Singing Performance Evaluation
Abstract
Automatic singing quality assessment (SQA) requires evaluating lyrical correctness and musical fidelity while handling expressive variations. However, existing systems largely rely on either acoustic cues or lyric transcriptions exclusively, limiting holistic performance evaluation. Furthermore, their integration is non-trivial due to challenges in robust singing transcription amid melisma, vibrato, and tempo elasticity. To this end, we propose MusicJudge, a modality-guided framework for automated SQA that performs block-aligned multimodal analysis by coupling lyric correctness with pitch-rhythm fidelity. It detects semantically meaningful lyric blocks using multi-signal matching that integrates semantic embeddings, lexical similarity, and phonetic alignment. To improve singing audio transcription, we introduce Modality-Guided LoRA for ASR fine-tuning. Experiments across datasets demonstrate strong agreement with human expert judgments and validate the generalizability of MusicJudge.
Abstract (translated)
URL
https://arxiv.org/abs/2606.26451
PDF
-
FoleySet: A Multi-Level Human-Annotated Foley Sound Dataset
Abstract
In audiovisual post-production, Foley refers to synchronous sound effects associated with human actions, such as footsteps, cloth rustle, and prop handling, that are recreated to match the on-screen movements and interactions of characters. These sounds are often recorded by professional Foley artists using physical props. This resource-intensive workflow has motivated data-driven research on Foley, including tasks such as classification, retrieval, and generation; however, high-quality annotated Foley datasets for training remain scarce. To address this gap, we present FoleySet, a publicly available Foley dataset of 10,000 audio clips annotated with a two-level Foley taxonomy. This dataset provides a standardized, Creative Commons-licensed resource for data-driven Foley classification, retrieval, and generation.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25980
PDF
-
Enhancing BEST-RQ Pseudo-Label Quality through Online Refinement for Automatic Speech Recognition
2026-06-24 15:32:47Jingjing Xu, Zijian Yang, Mohammad Zeineldeen, Eugen Beck, Ralf Schlueter, Hermann NeyAbstract
BEST-RQ is a simple and effective self-supervised training method for speech representation learning that performs well on automatic speech recognition (ASR) tasks. It generates pseudolabels using a fixed online quantization scheme, which simplifies training but provides weaker supervision than HuBERT-style models that iteratively refine pseudo-labels. In this work, we improve online pseudo-label generation while preserving simplicity. We propose three modifications: replacing the quantizer's linear projection with Principal Component Analysis (PCA), updating the codebook via iterative codebook refinement, and introducing an additional codebook updated via codebook distillation. We pre-train on the LibriSpeech 960-hour dataset and fine-tune using 100 hours of supervised LibriSpeech data. With all three modifications enabled, we achieve a 12% relative reduction in word error rate (WER) on the LibriSpeech test-other set, improving from 10.1% to 8.8%.
Abstract (translated)
URL
https://arxiv.org/abs/2606.30671
PDF
-
Soroll-IA: A Weakly Labeled Audio Dataset for Real-World Industrial Port Monitoring
2026-06-24 15:05:09Javier Naranjo-Alcazar, Jordi Grau-Haro, Ruben Ribes-Serrano, Marta Garcia-Ballesteros, Pedro ZuccarelloAbstract
Soroll-IA is a weakly labeled environmental audio dataset recorded in a real-world industrial port environment in Valencia (Spain) using two fixed sensing nodes. The dataset comprises approximately 22 hours of audio segmented into 7,396 clips and covers 26 sound event classes representative of industrial port acoustic activity commonly observed in such environments, such as crane sirens, train movements, traffic, and other logistical and industrial sounds. Recordings were captured under highly challenging acoustic conditions, including strong background noise, long-distance sources, and frequent event overlap. All audio clips were annotated by domain experts following a weak labeling strategy, where tags indicate the presence of sound events within a clip without temporal localization. To account for inter-annotator variability, two ground-truth versions are released: one without cross-validation, where a class is considered present if annotated by at least one expert, and a second, more conservative version based on cross-validation, where agreement by at least two-thirds of the annotators is required. The dataset is intended to support research in audio tagging, weakly supervised sound event detection, and machine learning under realistic industrial acoustic conditions. Benchmark results are provided using two complementary architectures: CNN14 representing high-capacity convolutional models for audio tagging, and MobileNetV2, selected for its suitability in real-time classification on low-resource edge devices. To the best of current knowledge, Soroll-IA constitutes an available dataset dedicated exclusively to industrial port acoustic environments, aiming to foster advances in robust environmental sound analysis for safety-critical and operational monitoring applications. The dataset is available online and collected under Attribution-NonCommercial 4.0 International license.
Abstract (translated)
URL
https://arxiv.org/abs/2606.26195
PDF
-
InstructFX2FX: A Multi-turn Text-to-Preset Demo for Iterative Audio Effect Refinement
2026-06-24 14:11:18Song-Ze Yu, Milan Liessens Dujardin, Yuxuan Cai, Wantong ZhangAbstract
We present InstructFX2FX, an interactive demonstration of multi-turn, text-guided audio effect refinement. Existing text-to-preset systems are largely single-shot, mapping one textual descriptor to one preset. Real audio engineering is instead sequential: engineers refine an existing effect chain through successive instructions such as "make it warmer" or "too harsh, soften it." This poses a stateful problem that single-shot systems do not address - given the current effect parameters and a new instruction, update the sound while preserving what earlier instructions already achieved. InstructFX2FX tackles this with a hybrid architecture that divides labor between a language model and CLAP-guided optimization. The LLM acts as a high-level planner that selects effects, orders the signal chain, and proposes the initial parameter state, leveraging recent evidence that LLMs can outperform CLAP-based optimization for single-turn text-to-parameter mapping; CLAP-guided optimization then refines the existing parameter state from this initialization, avoiding the parameter drift caused by re-prompting the LLM at each turn. At the demo, attendees drive a dry recording through successive natural-language instructions while staying in the loop: after each turn they audition the saved optimization checkpoints to choose how strongly an effect is applied, then issue the next instruction based on what still differs from the sound they intend. On SocialFX-derived descriptor transitions, CLAP-guided refinement lowers target-directed DSP-feature MMD on 9 of 10 directed descriptor pairs relative to an LLM-only reprompting baseline, with most refinement trajectories trending toward the intended target.
Abstract (translated)
URL
https://arxiv.org/abs/2606.22005
PDF
-
Frequency-Aware Self-Supervised Music Representation Learning
2026-06-24 11:26:38Yicheng Gu, Junan Zhang, Jerry Li, Zhizheng Wu, Lauri JuvelaAbstract
Self-supervised learning (SSL) has emerged as an essential paradigm for music information retrieval (MIR). While current SSL models achieve state-of-the-art performance across various MIR tasks, they typically treat audio as 1D sequences, either operating on time-domain waveforms or on flattened time-frequency-domain spectrograms. This discards the rich spatial and structural information in time-frequency representations and overlooks a fundamental intuition in music production. In particular, music is naturally represented as time-frequency grids in MIDI-based workflows, a structure that tightly corresponds to 2D spectrograms and inherently makes many MIR tasks trivial. Motivated by this intuition, we propose PupuJEPA, a visual Joint-Embedding Predictive Architecture (JEPA) that is trained directly on 2D spectrograms. Instead of applying masked language modeling (MLM) to 1D sequences, PupuJEPA learns robust representations by predicting the latent embeddings of masked 2D spectrogram patches from unmasked contexts. To optimally adapt such a visual framework to music signals, we also apply domain-specific modifications to model architecture, training scheme, and inference paradigm, with comprehensive ablation studies showing their effectiveness. Evaluations on the MARBLE benchmark show that PupuJEPA outperforms the 1D sequence-based SSL models across multiple MIR tasks in linear probing. Additionally, case studies of the attention maps also confirm that PupuJEPA captures musically meaningful patterns within the 2D time-frequency domain. Codes and checkpoints are available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25713
PDF
-
Joint Residual Reweighting for Classifier Free Guidance in Flow-Matching Zero-Shot TTS
2026-06-24 10:34:27Runwu Shi, Yujin Wang, Hongjin Song, Chunxiang JinAbstract
Classifier-free guidance (CFG) is widely used in flow-matching-based zero-shot text-to-speech (TTS), where generation is typically controlled by two conditions: the target text and a prompt speech signal. Standard CFG strengthens these conditions jointly, while recent branch-selective guidance methods attempt to enhance text or speaker conditioning separately, often leading to a trade-off between text correctness and speaker similarity. In this paper, we revisit the CFG under independently masked text and speech-prompt conditions, and decompose the guidance field into text, speaker, and joint residuals. We show that conventional speaker-selective guidance entangles the speaker residual with the joint residual, which may disturb text-related generation. Based on this observation, we propose joint residual reweighting, which independently controls the speaker and joint residuals within the standard CFG framework. Experiments on F5-TTS and CosyVoice2 show that the proposed method improves speaker similarity while maintaining competitive text correctness, demonstrating the usefulness of the joint residual for balancing speaker fidelity and text accuracy in zero-shot TTS.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25672
PDF
-
One Model, Many Latencies: Universal Speech Enhancement for Diverse Real-Time Applications
2026-06-24 09:28:55Szu-Wei Fu, Rong Chao, Xuesong Yang, Sung-Feng Huang, Ante Juki\'c, Yu Tsao, Yu-Chiang Frank WangAbstract
Different real-time speech applications impose distinct latency budgets, often requiring separately trained enhancement models for each scenario. In this paper, we propose a one-for-all, real-time universal speech enhancement model that provides explicit control over both algorithmic and computational latency. Algorithmic latency is flexibly adjusted via configurable look-ahead frames. To avoid learning inefficiency caused by varying padding configurations, we introduce parallel convolutional layers corresponding to different look-ahead settings. Computational latency is controlled through an early-exit mechanism, enabling inference at different network depths. To narrow the performance gap between specialized and flexible models, we propose a two-stage training strategy with a shared-to-multiple decoder transition. Overall, the proposed framework enables a single model to be deployed across diverse latency budgets without retraining separate models.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25621
PDF
-
STEB: A Speech-to-Speech Translation Expressiveness Benchmark for Evaluating Beyond Translation Fidelity
2026-06-24 08:06:41Sitong Cheng, Weizhen Bian, Songjun Cao, Jin Li, Bei Liu, Chunyang Jiang, Yike Zhang, Weihao Wu, Yiming Li, Chi-Min Chan, Long Ma, Wei XueAbstract
Speech-to-speech translation (S2ST) should preserve not only lexical meaning, but also expressive attributes: emotion, scenario style (e.g., news reporting vs. dramatic dialogue), and nonverbal vocalizations (NVs). Moreover, collecting cross-lingual target speech that is both translation-faithful and expressively aligned with the source is difficult at scale, making reference-based evaluation impractical. We introduce STEB (Speech-to-Speech Translation Expressiveness Benchmark), a 32.6-hour Chinese--English benchmark that evaluates both standard dimensions (translation fidelity, speaker similarity, duration alignment) and expressiveness dimensions (emotion, scenario style, NV preservation). For expressiveness evaluation, STEB uses a caption-then-summarize framework that converts speech into structured expressive attributes and compares source and hypothesis attributes with an LLM judge. Human validation shows statistically significant correlations with listener judgments across all expressive dimensions. We evaluate six S2ST systems covering cascaded systems, end-to-end models, and speech large language models. Many systems, especially cascaded ones, achieve strong translation fidelity, but they still struggle with emotion preservation (best: 3.82/5) and NV preservation (best: 2.31/5). These results reveal a gap between semantic transfer and expressive transfer, identifying expressiveness preservation as an open challenge for S2ST. Audio samples are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25529
PDF
-
CrossAccent-TTS: Cross-Lingual Accent-Intensity Controllable Text-to-Speech via Disentangled Speaker and Accent Representations
2026-06-24 05:02:36Ram Annamdevula, Ankit Tatawat, Ashishkumar P. Gudmalwar, Nirmesh J. Shah, Pankaj WasnikAbstract
Accent conversion and controllability remain fundamental challenges in cross-lingual text-to-speech (TTS), particularly for low-resource and phonetically diverse Indic languages. While recent large language model (LLM)-based TTS systems exhibit strong cross-lingual generalization, they provide limited explicit control over accent characteristics and intensity. In this paper, we propose CrossAccentTTS, a framework that enables both accent control and conversion while preserving speaker identity. Specifically, we introduce an Accent Intensity Controller (AIC) that injects weighted language embeddings into the accent subspace, allowing smooth interpolation between accents and fine-grained modulation of accent strength at inference time. Experiments on the Indic Multilingual and L2-arctic datasets shows that CrossAccent-TTS achieves precise control of accent intensity, outperforming strong baselines in accent similarity and controllability by maintaining speaker similarity and naturalness.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25403
PDF
-
From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models
2026-06-24 04:42:57Pengfei Zhang, Hoang H Nguyen, Kazi Shaharair Sharif, Yutong Song, Wenjun Huang, Henry Peng Zou, Pinxin Liu, Honghui Xu, Amir M. RahmaniAbstract
Recent Large Audio Language Models (LALMs) have achieved remarkable progress in audio perceptual tasks across individual acoustic layers, including speech, sound, and music. However, existing benchmarks predominantly evaluate these layers in isolation, overlooking the complex contextual relationships that arise when multiple acoustic sources co-occur in real-world auditory scenes. Real-world auditory interpretation requires Context-Aware Auditory Scene Understanding (CASU): the ability to comprehend the holistic scene by integrating sound layers. To evaluate this capability, we introduce the CASU benchmark, which assesses whether Audio LLMs can interpret auditory scenes composed of speech, acoustic events (e.g., announcements), and background environments (e.g., traffic), and reason about the logical relationships between these layers. We propose a scalable pipeline for constructing time-accurate, semi-synthetic audio streams by composing real-world scene sounds with synthetic speech. Building on this data, we design four tasks that probe scene understanding: contextual question answering, entity extraction from the scene, speaker role inference, and counterfactual reasoning where scene is manipulated. Experiments across multiple LALMs demonstrate that effective auditory scene understanding requires integration over all auditory layers, rather than reliance on speech or sound alone, underscoring the necessity of CASU for advancing complex audio understanding in LALMs.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25391
PDF
-
Supervised Post-training of Speech Foundation Models for Robust Adaptation in Speech Deepfake Detection
2026-06-24 02:49:17Zihan Pan, Sailor Hardik, Jinyang WuAbstract
Large speech foundation models have shown strong potential for speech deepfake detection, but direct fine-tuning is limited by a mismatch between self-supervised pre-training objectives and spoof-specific artifacts. To address this, we propose a mix-frame post-training strategy to create localized spoof-oriented perturbations and use frame-level supervision to encourage the SSL model to learn local inconsistencies that are critical for robust spoof detection. On ASVspoof5, we achieve state-of-the-art EER 4.50% for a single model without data augmentation. On ASVspoof2021 LA/DF, it further achieves only 0.16\% absolute EER gap between LA and DF, indicating strong and balanced robustness across distinct distortion conditions. These results show that supervised post-training provides an effective and practical way to adapt speech foundation models for robust deepfake detection.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25328
PDF
-
BCoughBench: Benchmarking Respiratory Acoustic Foundation Models Under Body-Coupled Wearable Sensor Conditions
2026-06-23 19:43:03Mayur Sanap, Prasanna Desikan, Edgar LobatonAbstract
Respiratory acoustic foundation models (FMs) are benchmarked exclusively on smartphone recordings, yet clinical deployment increasingly targets body-coupled (BC) wearables whose sensors attenuate high-frequency content through tissue and bone, leaving FM reliability uncharacterised. We introduce BCoughBench, evaluating five FMs (OPERA-CT/CE/GT, HeAR, M2D+Resp) on nine classification tasks (AUROC, sensitivity at 95% specificity, Expected Calibration Error) and three age regression tasks (MAE vs. a mean-predictor baseline) across five EBEN-simulated BC sensor conditions on five labeled cough datasets. Mean AUROC declines from 0.785 (smartphone) to 0.689-0.723, degrading most under temple vibration pickup ($\Delta$ = -0.096) and least under the soft in-ear ($\Delta$ = -0.062). No FM meets the clinical sensitivity threshold (Se@Sp95 $\geq$ 0.20) on most disease tasks under any BC sensor. Sex classification on the CIDRZ cohort collapses (AUROC 0.954 to 0.596-0.628, $\Delta$ = -0.341) while COVID detection is nearly unaffected ($\Delta$ = -0.004). Age regression is robust, improving under the forehead accelerometer on CoughVID (MAE 9.61 to 8.97 yr); HeAR leads on regression and demographic tasks, M2D+Resp on disease and characteristic tasks. BCoughBench provides a reproducible framework for FM evaluation under wearable conditions.
Abstract (translated)
URL
https://arxiv.org/abs/2606.25116
PDF
-
Statistical validation and full-sphere extension of a Bayesian model for human static sound localisation
2026-06-23 09:57:43Roberto Barumerli, Fabian Brinkmann, Emanuele Zanoni, Anton Hoyer, Lorenzo Picinali, Michele GeronazzoAbstract
Auditory models are central tools for studying spatial hearing, yet their validation typically relies on heuristic performance metrics rather than principled statistical methods. We present two contributions building on a Bayesian sound localisation model that jointly infers sound direction from noisy perceptual features and individual head-related transfer functions (HRTFs). First, we derive an explicit likelihood function and validate it through parameter recovery on simulated data and fitting to behavioural responses from 33 participants, demonstrating that the framework reliably identifies individual sensorimotor and spectral parameters. Second, we use this framework to compare four HRTF template interpolation methods, showing that full-sphere spatial coverage and high-frequency spectral fidelity are the primary determinants of template quality, while the specific interpolation algorithm is secondary. Together, these results show that standard model-based statistical methods can address both fundamental questions in spatial hearing and applied problems such as perceptual HRTF evaluation. An open-source Python implementation is released alongside this work.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24367
PDF
-
The effect of micro-changes in the pluck trajectory on the sound of an acoustic guitar
2026-06-23 09:43:48Marek Pluta, Jan Jasi\'nski, Daniel Tokarczyk, Julia GrygielAbstract
This study explores how micro-changes in the plucking trajectory of a guitar pick influence the sound of an acoustic guitar. Using a state-of-the-art robotic plucker, a series of measurements has been performed, where the plectrum was moved towards the instrument by a step of 192 micrometers, resulting in an increased attack depth. It has been analysed how the effect of these changes is reflected in loudness, timbre, harmonic content and how the sound progresses during decay. This methodology has been repeated for guitar plectra made from six different materials to investigate how the pick itself influences the effect of a change in the plucking trajectory. The results of the study show that at a low depth the string is not fully excited resulting in weak and markedly altered sound. The range of this effect changes with the mechanical properties of the plectrum material. After this range an increase in depth results in an increase in sound loudness, a decrease in inharmonicity and noisiness and a shift in timbre where the sound becomes fuller in low frequencies and rougher. Presented findings help to understand the nuanced relationship between plucking trajectory and acoustic output. They provide important insights regarding the importance of plucking in guitar testing methodologies, showing that the mech
Abstract (translated)
URL
https://arxiv.org/abs/2606.24356
PDF
-
What Does a Pathological Speech Assessment Model Know about Acoustic Features? A Case Study on Oral and Oropharyngeal Cancer Patients
2026-06-23 07:37:10Tuan Nguyen (LIA, AU), Corinne Fredouille (AU, LIA), Alain Ghio (LPL), Muriel Lalain (LPL), Virginie Woisard (UT2J, UT3, LNPL)Abstract
This work investigates the interpretability of a Wav2Vec 2.0based speech intelligibility assessment model for oral and oropharyngeal cancer patients through canonical correlation analysis. By measuring the correlation between the model embeddings and eGeMAPS low-level descriptors (LLDs) as an interpretable reference, we analyze how acoustic information is encoded across the model layers. The analysis is conducted at two levels: individual LLDs layer-wise, and group-level: prosodic, spectral, and voice quality. Results show that the learned representations are most strongly correlated with spectral and prosodic features, with the first MFCC coefficient yielding the highest correlations across all layers. At the group level, spectral and prosodic groups achieve correlations of 0.77 and 0.71 respectively, while voice quality reaches 0.65. Beyond model interpretability, this work also offers practical guidance on acoustic feature selection for pathological speech assessment.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24949
PDF
-
Evaluation of Headrest-Integrated Loudspeakers for Enhanced Spatial Audio Immersion in Automotive Cabins
2026-06-23 05:02:54Martin Wolters, Jacobo Giralt, Harald Mundt, Arijit BiswasAbstract
Immersive object-based spatial audio is now firmly established in the music industry as the standard for production, distribution, and playback. The number of automobiles integrating such content to provide premium entertainment experiences is steadily increasing, driving the development of new audio rendering techniques. While loudspeakers integrated into automotive headrests have been around for more than 50 years, they have not yet achieved status as a standard feature in new cars. However, they represent a powerful tool for reproducing immersive audio by enabling the creation of personal sound zones with reduced passenger distraction while effectively complementing existing cabin speakers. We conducted subjective assessments using paired comparison experiments to measure preference and multiple spatial audio attributes. We modeled the resulting probability outcomes using a probabilistic choice model, the Bradley-Terry-Luce rank ordering. The results indicate that headrest-integrated speakers can improve the audio perception in immersive audio scenarios.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24146
PDF
-
Joint Learning of Covariance Estimation and White Noise Gain for Robust MVDR Beamforming
2026-06-23 04:44:54Yongyi Deng, Hanchen Pei, Jianbo Ma, Gongping Huang, Jingdong Chen, Jacob BenestyAbstract
The minimum variance distortionless response (MVDR) beamformer is widely used for multichannel speech enhancement due to strong noise suppression while preserving target signals. In practice, its performance is sensitive to microphone self-noise and array mismatches. Existing approaches typically rely on fixed, manually tuned WNG thresholds or diagonal loading, leading to suboptimal performance under unknown or time-varying acoustic conditions. This paper proposes a data-driven MVDR framework that adaptively estimates the WNG constraint using a deep neural network. The network jointly predicts a time-frequency noise mask for covariance estimation and a frequency-dependent WNG threshold, enabling dynamic robustness-directivity control. A differentiable robust MVDR layer is integrated into the framework, allowing end-to-end optimization. Experiments demonstrate consistent improvements in speech quality and intelligibility over conventional fixed-WNG MVDR methods.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24137
PDF
-
Aligning MusicLLM with Emotion using Instruction Tuning and Feedback-Driven Alignment
Abstract
This paper investigates whether music large language models (MusicLLMs) can be aligned for emotion regression. While MusicLLMs have shown strong performance in music information retrieval tasks, their ability to predict arousal and valence scores remains limited, since emotion regression has not been an explicit training objective. To examine whether MusicLLMs can be aligned with emotion, we train MusicLLMs on emotion regression and compare two strategies: instruction tuning and feedback-driven alignment. Our experiments show that task-aware instruction tuning enables MusicLLMs to predict emotion levels to some extent, although the accuracy remains limited. Applying feedback-driven alignment with a verifiable numerical reward substantially improves performance on both arousal and valence over instruction tuning alone. We further show that our approach improves emotion regression performance while maintaining MusicQA capability.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24123
PDF
-
A Fusion-Aware Two-Stage Framework for Mispronunciation Detection and Diagnosis in Low-Resource Modern Standard Arabic
2026-06-23 03:02:00Jing Yang, Shuqing Zhang, Yongyi Deng, Pan Li, Ting Dang, Gongping Huang, Jingdong Chen, Jacob BenestyAbstract
Accurate phoneme recognition is pivotal for mispronunciation detection and diagnosis (MDD) in modern standard Arabic (MSA), yet remains constrained by data scarcity and the synthetic-real domain gap. This work proposes a two-stage end-to-end framework. It integrates a pre-trained encoder with causal dilated temporal convolutional networks to preserve fine-grained phonetic variations. A hierarchical two-stage strategy first learns general mappings from native/synthetic corpora, then adapts to scarce real learner data to mitigate domain shift without over-correction. Prediction stability is further enhanced via multi-checkpoint ensemble inference with N-gram rescoring. Evaluated on the QuranMB.v2 test set, our system achieves an F1-score of $0.7201$, a $63.1$\% relative improvement over baseline ($0.4414$). This performance ranks at the top of the IqraEval.2 Challenge, establishing a new state-of-the-art for low-resource MSA in MDD.
Abstract (translated)
URL
https://arxiv.org/abs/2606.24086
PDF
