arXiv subjects

Sabato Marco Siniscalchi

Publications and source records attributed to Sabato Marco Siniscalchi.

At least 19 recordsLinked to original sources

arXiv2026

RT-SEMamba: Real-Time Speech Enhancement Mamba via Progressive Knowledge Distillation

We present RT-SEMamba, a fully causal speech enhancement (SE) model built upon causal time-frequency Mamba blocks. Unlike Transformer-based architectures that rely on a growing key-value cache, Mamba propagates a fixed-size recurrent state per layer, enabling memory- and bandwidth-efficient long-form inference. We further introduce a progressive knowledge distillation (KD) strategy that compresses an 8-layer teacher into a shallow 1-layer student by jointly distilling complex spectral outputs and intermediate representations. On Voicebank-DEMAND, the 8-layer RT-SEMamba achieves 3.32 PESQ with a 25 ms algorithmic latency constraint, and the distilled 1-layer student improves over a naive 1-layer baseline from 3.06 to 3.18 PESQ while preserving the same steady-state RTF, delivering a 2.64x speedup over the teacher. These results demonstrate that state-space models with progressive KD provide a competitive quality-latency trade-off for real-time SE.

cs.SD↗

arXiv2026

A Knowledge-Driven Approach to Target Speech Extraction in the Presence of Background Sound Effects for Cinematic Audio Source Separation (CASS)

We propose a knowledge-driven approach to speech target extraction in the presence of background sound effects already recorded in cinematic audio. The specific knowledge sources studied are manners of articulation that are detected in speech frames and adopted to form a knowledge vector as a part of features to enhance speech separation and target speech extraction because some short speech segments are often difficult to separate from mixed background sounds. Testing on the recent Sound Demixing Challenge data for cinematic audio source separation (CASS) shows that utilizing articulator-aware knowledge sources produces better separation results than those obtained without using any knowledge, especially for speech segments buried in unspecified background sound events.

eess.AS↗

arXiv2026

Few-Shot and Pseudo-Label Guided Speech Quality Evaluation with Large Language Models

In this paper, we introduce GatherMOS, a novel framework that leverages large language models (LLM) as meta-evaluators to aggregate diverse signals into quality predictions. GatherMOS integrates lightweight acoustic descriptors with pseudo-labels from DNSMOS and VQScore, enabling the LLM to reason over heterogeneous inputs and infer perceptual mean opinion scores (MOS). We further explore both zero-shot and few-shot in-context learning setups, showing that zero-shot GatherMOS maintains stable performance across diverse conditions, while few-shot guidance yields large gains when support samples match the test conditions. Experiments on the VoiceBank-DEMAND dataset demonstrate that GatherMOS consistently outperforms DNSMOS, VQScore, naive score averaging, and even learning-based models such as CNN-BLSTM and MOS-SSL when trained under limited labeled-data conditions. These results highlight the potential of LLM-based aggregation as a practical strategy for non-intrusive speech quality evaluation.

eess.AS↗

arXiv2026

A Knowledge-Driven Approach to Music Segmentation, Music Source Separation and Cinematic Audio Source Separation

We propose a knowledge-driven, model-based approach to segmenting audio into single-category and mixed-category chunks with applications to source separation. "Knowledge" here denotes information associated with the data, such as music scores. "Model" here refers to tool that can be used for audio segmentation and recognition, such as hidden Markov models. In contrast to conventional learning that often relies on annotated data with given segment categories and their corresponding boundaries to guide the learning process, the proposed framework does not depend on any pre-segmented training data and learns directly from the input audio and its related knowledge sources to build all necessary models autonomously. Evaluation on simulation data shows that score-guided learning achieves very good music segmentation and separation results. Tested on movie track data for cinematic audio source separation also shows that utilizing sound category knowledge achieves better separation results than those obtained with data-driven techniques without using such information.

eess.AS↗

arXiv2026

TimeOmni-1: Incentivizing Complex Reasoning with Time Series in Large Language Models

Recent advances in multimodal time series learning underscore a paradigm shift from analytics centered on basic patterns toward advanced time series understanding and reasoning. However, existing multimodal time series datasets mostly remain at the level of surface alignment and question answering, without reaching the depth of genuine reasoning. The absence of well-defined tasks that genuinely require time series reasoning, along with the scarcity of high-quality data, has limited progress in building practical time series reasoning models (TSRMs). To this end, we introduce Time Series Reasoning Suite (TSR-Suite), which formalizes four atomic tasks that span three fundamental capabilities for reasoning with time series: (1) perception, acquired through scenario understanding and causality discovery; (2) extrapolation, realized via event-aware forecasting; and (3) decision-making, developed through deliberation over perception and extrapolation. TSR-Suite is the first comprehensive time series reasoning suite that supports not only thorough evaluation but also the data pipeline and training of TSRMs. It contains more than 23K samples, of which 2.3K are carefully curated through a human-guided hierarchical annotation process. Building on this foundation, we introduce TimeOmni-1, the first unified reasoning model designed to address diverse real-world problems demanding time series reasoning. The model is trained in multiple stages, integrating a mixture of task scenarios, novel reward functions, and tailored optimizations. Experiments show that TimeOmni-1 delivers strong out-of-distribution generalization across all tasks and achieves a high rate of valid responses. It significantly improves causality discovery accuracy (64.0% vs. 35.9% with GPT-4.1) and raises the valid response rate by over 6% compared to GPT-4.1 on the event-aware forecasting task.

cs.LG↗

arXiv2026

MDM-ASR: Bridging Accuracy and Efficiency in ASR with Diffusion-Based Non-Autoregressive Decoding

In sequence-to-sequence Transformer ASR, autoregressive (AR) models achieve strong accuracy but suffer from slow decoding, while non-autoregressive (NAR) models enable parallel decoding at the cost of degraded performance. We propose a principled NAR ASR framework based on Masked Diffusion Models to reduce this gap. A pre-trained speech encoder is coupled with a Transformer diffusion decoder conditioned on acoustic features and partially masked transcripts for parallel token prediction. To mitigate the training-inference mismatch, we introduce Iterative Self-Correction Training that exposes the model to its own intermediate predictions. We also design a Position-Biased Entropy-Bounded Confidence-based sampler with positional bias to further boost results. Experiments across multiple benchmarks demonstrate consistent gains over prior NAR models and competitive performance with strong AR baselines, while retaining parallel decoding efficiency.

eess.AS↗

arXiv2026

Aligning Generative Speech Enhancement with Perceptual Feedback

Language Model (LM)-based speech enhancement (SE) has recently emerged as a promising direction, but existing approaches predominantly rely on token-level likelihood objectives that weakly reflect human perception. This mismatch limits progress, as optimizing signal accuracy does not always improve naturalness or listening comfort. We address this gap by introducing a perceptually aligned LM-based SE approach. Our method applies Direct Preference Optimization (DPO) with UTMOS, a neural MOS predictor, as a proxy for human ratings, directly steering models toward perceptually preferred outputs. This design directly connects model training to perceptual quality and is broadly applicable within LM-based SE frameworks. On the Deep Noise Suppression Challenge 2020 test sets, our approach consistently improves speech quality metrics, achieving relative gains of up to 56%. To our knowledge, this is the first integration of perceptual feedback into LM-based SE and the first application of DPO in the SE domain, establishing a new paradigm for perceptually aligned enhancement with SE.

eess.AS↗

arXiv2025

Towards Robust Assessment of Pathological Voices via Combined Low-Level Descriptors and Foundation Model Representations

Perceptual voice quality assessment plays a vital role in diagnosing and monitoring voice disorders. Traditional methods, such as the Consensus Auditory-Perceptual Evaluation of Voice (CAPE-V) and the Grade, Roughness, Breathiness, Asthenia, and Strain (GRBAS) scales, rely on expert raters and are prone to inter-rater variability, emphasizing the need for objective solutions. This study introduces the Voice Quality Assessment Network (VOQANet), a deep learning framework that employs an attention mechanism and Speech Foundation Model (SFM) embeddings to extract high-level features. To further enhance performance, we propose VOQANet+, which integrates self-supervised SFM embeddings with low-level acoustic descriptors-namely jitter, shimmer, and harmonics-to-noise ratio (HNR). Unlike previous approaches that focus solely on vowel-based phonation (PVQD-A), our models are evaluated on both vowel-level and sentence-level speech (PVQD-S) to assess generalizability. Experimental results demonstrate that sentence-based inputs yield higher accuracy, particularly at the patient level. Overall, VOQANet consistently outperforms baseline models in terms of root mean squared error (RMSE) and Pearson correlation coefficient across CAPE-V and GRBAS dimensions, with VOQANet+ achieving even greater performance gains. Additionally, VOQANet+ maintains consistent performance under noisy conditions, suggesting enhanced robustness for real-world and telehealth applications. This work highlights the value of combining SFM embeddings with low-level features for accurate and robust pathological voice assessment.

cs.SD↗

arXiv2025

Hallucination Benchmark for Speech Foundation Models

Hallucinations in automatic speech recognition (ASR) systems refer to fluent and coherent transcriptions produced by neural ASR models that are completely unrelated to the underlying acoustic input (i.e., the speech signal). While similar to conventional decoding errors in potentially compromising the usability of transcriptions for downstream applications, hallucinations can be more detrimental due to their preservation of syntactically and semantically plausible structure. This apparent coherence can mislead subsequent processing stages and introduce serious risks, particularly in critical domains such as healthcare and law. Conventional evaluation metrics are primarily centered on error-based metrics and fail to distinguish between phonetic inaccuracies and hallucinations. Consequently, there is a critical need for new evaluation frameworks that can effectively identify and assess models with a heightened propensity for generating hallucinated content. To this end, we introduce SHALLOW, the first benchmark framework that systematically categorizes and quantifies hallucination phenomena in ASR along four complementary axes: lexical, phonetic, morphological, and semantic. We define targeted metrics within each category to produce interpretable profiles of model behavior. Through evaluation across various architectures and speech domains, we have found that SHALLOW metrics correlate strongly with word error rate (WER) when recognition quality is high (i.e., low WER). Still, this correlation weakens substantially as WER increases. SHALLOW, therefore, captures fine-grained error patterns that WER fails to distinguish under degraded and challenging conditions. Our framework supports specific diagnosis of model weaknesses and provides feedback for model improvement beyond what aggregate error rates can offer.

cs.CL↗

arXiv2025

An Investigation of Incorporating Mamba for Speech Enhancement

This work aims to investigate the use of a recently proposed, attention-free, scalable state-space model (SSM), Mamba, for the speech enhancement (SE) task. In particular, we employ Mamba to deploy different regression-based SE models (SEMamba) with different configurations, namely basic, advanced, causal, and non-causal. Furthermore, loss functions either based on signal-level distances or metric-oriented are considered. Experimental evidence shows that SEMamba attains a competitive PESQ of 3.55 on the VoiceBank-DEMAND dataset with the advanced, non-causal configuration. A new state-of-the-art PESQ of 3.69 is also reported when SEMamba is combined with Perceptual Contrast Stretching (PCS). Compared against Transformed-based equivalent SE solutions, a noticeable FLOPs reduction up to ~12% is observed with the advanced non-causal configurations. Finally, SEMamba can be used as a pre-processing step before automatic speech recognition (ASR), showing competitive performance against recent SE solutions.

cs.SD↗

arXiv2025

HDA-SELD: Hierarchical Cross-Modal Distillation with Multi-Level Data Augmentation for Low-Resource Audio-Visual Sound Event Localization and Detection

This work presents HDA-SELD, a unified framework that combines hierarchical cross-modal distillation (HCMD) and multi-level data augmentation to address low-resource audio-visual (AV) sound event localization and detection (SELD). An audio-only SELD model acts as the teacher, transferring knowledge to an AV student model through both output responses and intermediate feature representations. To enhance learning, data augmentation is applied by mixing features randomly selected from multiple network layers and associated loss functions tailored to the SELD task. Extensive experiments on the DCASE 2023 and 2024 Challenge SELD datasets show that the proposed method significantly improves AV SELD performance, yielding relative gains of 21%-38% in the overall metric over the baselines. Notably, our proposed HDA-SELD achieves results comparable to or better than teacher models trained on much larger datasets, surpassing state-of-the-art methods on both DCASE 2023 and 2024 Challenge SELD tasks.

cs.SD↗

arXiv2025

A Bottom-up Framework with Language-universal Speech Attribute Modeling for Syllable-based ASR

We propose a bottom-up framework for automatic speech recognition (ASR) in syllable-based languages by unifying language-universal articulatory attribute modeling with syllable-level prediction. The system first recognizes sequences or lattices of articulatory attributes that serve as a language-universal, interpretable representation of pronunciation, and then transforms them into syllables through a structured knowledge integration process. We introduce two evaluation metrics, namely Pronunciation Error Rate (PrER) and Syllable Homonym Error Rate (SHER), to evaluate the model's ability to capture pronunciation and handle syllable ambiguities. Experimental results on the AISHELL-1 Mandarin corpus demonstrate that the proposed bottom-up framework achieves competitive performance and exhibits better robustness under low-resource conditions compared to the direct syllable prediction model. Furthermore, we investigate the zero-shot cross-lingual transferability on Japanese and demonstrate significant improvements over character- and phoneme-based baselines by 40% error rate reduction.

eess.AS↗

arXiv2025

An Investigation on Combining Geometry and Consistency Constraints into Phase Estimation for Speech Enhancement

We propose a novel iterative phase estimation framework, termed multi-source Griffin-Lim algorithm (MSGLA), for speech enhancement (SE) under additive noise conditions. The core idea is to leverage the ad-hoc consistency constraint of complex-valued short-time Fourier transform (STFT) spectrograms to address the sign ambiguity challenge commonly encountered in geometry-based phase estimation. Furthermore, we introduce a variant of the geometric constraint framework based on the law of sines and cosines, formulating a new phase reconstruction algorithm using noise phase estimates. We first validate the proposed technique through a series of oracle experiments, demonstrating its effectiveness under ideal conditions. We then evaluate its performance on the VB-DMD and WSJ0-CHiME3 data sets, and show that the proposed MSGLA variants match well or slightly outperform existing algorithms, including direct phase estimation and DNN-based sign prediction, especially in terms of background noise suppression.

eess.AS↗

arXiv2025

Joint Tensor-Train Parameterization for Efficient and Expressive Low-Rank Adaptation

Low-Rank Adaptation (LoRA) is widely recognized for its parameter-efficient fine-tuning of large-scale neural models. However, standard LoRA independently optimizes low-rank matrices, which inherently limits its expressivity and generalization capabilities. While classical tensor-train (TT) decomposition can be separately employed on individual LoRA matrices, this work demonstrates that the classical TT-based approach neither significantly improves parameter efficiency nor achieves substantial performance gains. This paper proposes TensorGuide, a novel tensor-train-guided adaptation framework to overcome these limitations. TensorGuide generates two correlated low-rank LoRA matrices through a unified TT structure driven by controlled Gaussian noise. The resulting joint TT representation inherently provides structured, low-rank adaptations, significantly enhancing expressivity, generalization, and parameter efficiency without increasing the number of trainable parameters. Theoretically, we justify these improvements through neural tangent kernel analyses, demonstrating superior optimization dynamics and enhanced generalization. Extensive experiments on quantum dot classification and GPT-2 fine-tuning benchmarks demonstrate that TensorGuide-based LoRA consistently outperforms standard LoRA and TT-LoRA, achieving improved accuracy and scalability with fewer parameters.

cs.LG↗

arXiv2025

MVP: Multi-source Voice Pathology detection

Voice disorders significantly impact patient quality of life, yet non-invasive automated diagnosis remains under-explored due to both the scarcity of pathological voice data, and the variability in recording sources. This work introduces MVP (Multi-source Voice Pathology detection), a novel approach that leverages transformers operating directly on raw voice signals. We explore three fusion strategies to combine sentence reading and sustained vowel recordings: waveform concatenation, intermediate feature fusion, and decision-level combination. Empirical validation across the German, Portuguese, and Italian languages shows that intermediate feature fusion using transformers best captures the complementary characteristics of both recording types. Our approach achieves up to +13% AUC improvement over single-source methods.

eess.AS↗

arXiv2025

Exploring Generative Error Correction for Dysarthric Speech Recognition

Despite the remarkable progress in end-to-end Automatic Speech Recognition (ASR) engines, accurately transcribing dysarthric speech remains a major challenge. In this work, we proposed a two-stage framework for the Speech Accessibility Project Challenge at INTERSPEECH 2025, which combines cutting-edge speech recognition models with LLM-based generative error correction (GER). We assess different configurations of model scales and training strategies, incorporating specific hypothesis selection to improve transcription accuracy. Experiments on the Speech Accessibility Project dataset demonstrate the strength of our approach on structured and spontaneous speech, while highlighting challenges in single-word recognition. Through comprehensive analysis, we provide insights into the complementary roles of acoustic and linguistic modeling in dysarthric speech recognition

cs.CL↗

arXiv2025

"KAN you hear me?" Exploring Kolmogorov-Arnold Networks for Spoken Language Understanding

Kolmogorov-Arnold Networks (KANs) have recently emerged as a promising alternative to traditional neural architectures, yet their application to speech processing remains under explored. This work presents the first investigation of KANs for Spoken Language Understanding (SLU) tasks. We experiment with 2D-CNN models on two datasets, integrating KAN layers in five different configurations within the dense block. The best-performing setup, which places a KAN layer between two linear layers, is directly applied to transformer-based models and evaluated on five SLU datasets with increasing complexity. Our results show that KAN layers can effectively replace the linear layers, achieving comparable or superior performance in most cases. Finally, we provide insights into how KAN and linear layers on top of transformers differently attend to input regions of the raw waveforms.

cs.CL↗

arXiv2025

From KAN to GR-KAN: Advancing Speech Enhancement with KAN-Based Methodology

Deep neural network (DNN)-based speech enhancement (SE) usually uses conventional activation functions, which lack the expressiveness to capture complex multiscale structures needed for high-fidelity SE. Group-Rational KAN (GR-KAN), a variant of Kolmogorov-Arnold Networks (KAN), retains KAN's expressiveness while improving scalability on complex tasks. We adapt GR-KAN to existing DNN-based SE by replacing dense layers with GR-KAN layers in the time-frequency (T-F) domain MP-SENet and adapting GR-KAN's activations into the 1D CNN layers in the time-domain Demucs. Results on Voicebank-DEMAND show that GR-KAN requires up to 4x fewer parameters while improving PESQ by up to 0.1. In contrast, KAN, facing scalability issues, outperforms MLP on a small-scale signal modeling task but fails to improve MP-SENet. We demonstrate the first successful use of KAN-based methods for consistent improvement in both time- and SoTA TF-domain SE, establishing GR-KAN as a promising alternative for SE.

eess.AS↗