arXiv subjects

Saikat Chatterjee

Publications and source records attributed to Saikat Chatterjee.

At least 19 recordsLinked to original sources

arXiv2026

DNSMOS-C: Improving End-to-end Speech Quality Models via Contrastive Learning

We introduce DNSMOS-C, a compact end-to-end speech quality assessment model that extends the DNSMOS Pro framework by integrating a MOS-guided triplet-based contrastive loss. Applied directly to the intermediate embeddings, this contrastive supervision encourages the latent space to be better organized with respect to perceptual quality while preserving the simplicity and efficiency of DNSMOS Pro. Unlike prior methods that depend on large pre-trained self-supervised learning (SSL) encoders and multi-stage training, DNSMOS-C jointly learns speech representations and MOS regression within a single, unified framework. Experiments on multiple datasets show that DNSMOS-C consistently improves correlation metrics over DNSMOS Pro and achieves better generalization on challenging out-of-domain test sets. Furthermore, latent space analyses indicate that our approach learns representations that exhibit an emergent low-dimensional quality ordering, which enhances interpretability and improves training stability. These findings demonstrate that MOS-guided contrastive learning enables more robust and accurate quality predictions without incurring additional computational overhead.

eess.AS↗

arXiv2026

Generative Modeling for Physiological Signals

Physiological signals support clinical diagnosis, health monitoring, rehabilitation, wearable sensing, and human--machine interaction. However, their applications are often constrained by limited labeled data, class imbalance, noisy or incomplete recordings, heterogeneous acquisition settings, and privacy restrictions. Generative modeling has therefore attracted increasing attention as a means of addressing some of these barriers. Recent studies have used generative models to augment scarce datasets, restore degraded recordings, translate between modalities, and synthesize conditional physiological waveforms. This review summarizes recent work on generative modeling for cardiovascular, neural, muscular, peripheral, and specialized physiological signals. Major model families are covered, including generative adversarial networks (GANs), autoencoders and variational autoencoders (AEs/VAEs), diffusion models, autoregressive sequence models, and hybrid architectures. In addition, it organizes existing evaluation practices into a hierarchical framework spanning signal-level similarity, dataset-level distribution, physiological validity, task-oriented utility, and assessments of generalization and robustness. By linking signal-specific constraints, generative roles, model families, and evaluation evidence, this review provides structured guidance for the future use and evaluation of generative models in physiological-signal research.

eess.SP↗

arXiv2026

Diffusion-Based Heart Sound Generation: Evaluation with Physiological Signal Metrics, Classifiers, and Expert Listening

Publicly available phonocardiogram (PCG) datasets remain limited in size and pathological diversity, constraining both auscultation training and the generalisation of automated heart-sound classifiers. A class-conditional diffusion model for PCG generation is developed in the log-mel domain and synthetic fidelity is assessed using complementary (i) physiology-inspired plausibility metrics, (ii) downstream label-consistency evaluation, and (iii) expert listening. Experiments use the Phy-sioNet/Computing in Cardiology Challenge 2016 dataset (3240 recordings) with recording-level splits. After preprocessing and quality control, 16,749 non-overlapping 4 s clips are mapped to a normalised 1 x 128 x 128 log-mel representation to train a conditional 2D U-Net denoiser with classifier-free guidance. Signal-level plausibility is quantified on reconstructed waveforms using three lightweight metrics: an envelope-autocorrelation rhythm score, an amplitude-based explosion score, and the dominant cycle lag. Synthetic clips preserve similar dominant cycle durations but exhibit reduced envelope periodicity and increased transient burstiness relative to real clips. For downstream evaluation, a ResNet-50 classifier achieves 92.24% accuracy on the held-out real test set and 82.8% accuracy on class-balanced synthetic batches, indicating that generated signals retain discriminative structure relevant to normal/abnormal classification. In a pilot expert listening study (60 clips, two clinicians), most synthetic clips are judged as heart-sound-like, while abnormality sensitivity is low for both real and synthetic 4 s excerpts. Overall, the results provide a practical baseline for diffusion-based PCG generation while highlighting remaining challenges in retaining abnormal acoustic cues and reducing reconstruction-induced artefacts.

eess.SP↗

arXiv2026

SA-SSL-MOS: Self-supervised Learning MOS Prediction with Spectral Augmentation for Generalized Multi-Rate Speech Assessment

Designing a speech quality assessment (SQA) system for estimating mean-opinion-score (MOS) of multi-rate speech with varying sampling frequency (16-48 kHz) is a challenging task. The challenge arises due to the limited availability of a MOS-labeled training dataset comprising multi-rate speech samples. While self-supervised learning (SSL) models have been widely adopted in SQA to boost performance, a key limitation is that they are pretrained on 16 kHz speech and therefore discard high-frequency information present in higher sampling rates. To address this issue, we propose a spectrogram-augmented SSL method that incorporates high-frequency features (up to 48 kHz sampling rate) through a parallel-branch architecture. We further introduce a two-step training scheme: the model is first pre-trained on a large 48 kHz dataset and then fine-tuned on a smaller multi-rate dataset. Experimental results show that leveraging high-frequency information overlooked by SSL features is crucial for accurate multi-rate SQA, and that the proposed two-step training substantially improves generalization when multi-rate data is limited.

eess.AS↗

arXiv2026

DNS: Data-driven Nonlinear Smoother for Complex Model-free Process

We propose data-driven nonlinear smoother (DNS) to estimate a hidden state sequence of a complex dynamical process from a noisy, linear measurement sequence. The dynamical process is model-free, that is, we do not have any knowledge of the nonlinear dynamics of the complex process. There is no state-transition model (STM) of the process available. The proposed DNS uses a recurrent architecture that helps to provide a closed-form posterior of the hidden state sequence given the measurement sequence. DNS learns in an unsupervised manner, meaning the training dataset consists of only measurement data and no state data. We demonstrate DNS using simulations for smoothing of several stochastic dynamical processes, including a benchmark Lorenz system. Experimental results show that the DNS is significantly better than a deep Kalman smoother (DKS) and an iterative data-driven nonlinear state estimation (iDANSE) smoother.

eess.SP↗

arXiv2026

VSE: Variational state estimation of complex model-free process

We design a variational state estimation (VSE) method that provides a closed-form Gaussian posterior of an underlying complex dynamical process from (noisy) nonlinear measurements. The complex process is model-free. That is, we do not have a suitable physics-based model characterizing the temporal evolution of the process state. The closed-form Gaussian posterior is provided by a recurrent neural network (RNN). The use of RNN is computationally simple in the inference phase. For learning the RNN, an additional RNN is used in the learning phase. Both RNNs help each other learn better based on variational inference principles. The VSE is demonstrated for a tracking application - state estimation of a stochastic Lorenz system (a benchmark process) using a 2-D camera measurement model. The VSE is shown to be competitive against a particle filter that knows the Lorenz system model and a recently proposed data-driven state estimation method that does not know the Lorenz system model.

eess.SP↗

arXiv2025

pDANSE: Particle-based Data-driven Nonlinear State Estimation from Nonlinear Measurements

We consider the problem of designing a data-driven nonlinear state estimation (DANSE) method that uses (noisy) nonlinear measurements of a process whose underlying state transition model (STM) is unknown. Such a process is referred to as a model-free process. A recurrent neural network (RNN) provides parameters of a Gaussian prior that characterize the state of the model-free process, using all previous measurements at a given time point. In the case of DANSE, the measurement system was linear, leading to a closed-form solution for the state posterior. However, the presence of a nonlinear measurement system renders a closed-form solution infeasible. Instead, the secondorder statistics of the state posterior are computed using the nonlinear measurements observed at the time point. We address the nonlinear measurements using a reparameterization trickbased particle sampling approach, and estimate the second-order statistics of the state posterior. The proposed method is referred to as particle-based DANSE (pDANSE). The RNN of pDANSE uses sequential measurements efficiently and avoids the use of computationally intensive sequential Monte-Carlo (SMC) and/or ancestral sampling. We describe the semi-supervised learning method for pDANSE, which transitions to unsupervised learning in the absence of labeled data. Using a stochastic Lorenz-63 system as a benchmark process, we experimentally demonstrate the state estimation performance for four nonlinear measurement systems. We explore cubic nonlinearity and a cameramodel nonlinearity where unsupervised learning is used; then we explore half-wave rectification nonlinearity and Cartesian-tospherical nonlinearity where semi-supervised learning is used. Additionally, we also show the performance of pDANSE for the stochastic Lorenz-96 system with a half-wave, rectified measurement system. The performance of state estimation is shown to be competitive vis-a-vis model-driven methods that have complete knowledge of the STM of the dynamical system.

eess.SP↗

arXiv2025

Selection of Layers from Self-supervised Learning Models for Predicting Mean-Opinion-Score of Speech

Self-supervised learning (SSL) models like Wav2Vec2, HuBERT, and WavLM have been widely used in speech processing. These transformer-based models consist of multiple layers, each capturing different levels of representation. While prior studies explored their layer-wise representations for efficiency and performance, speech quality assessment (SQA) models predominantly rely on last-layer features, leaving intermediate layers underexamined. In this work, we systematically evaluate different layers of multiple SSL models for predicting mean-opinion-score (MOS). Features from each layer are fed into a lightweight regression network to assess effectiveness. Our experiments consistently show early-layers features outperform or match those from the last layer, leading to significant improvements over conventional approaches and state-of-the-art MOS prediction models. These findings highlight the advantages of early-layer selection, offering enhanced performance and reduced system complexity.

eess.AS↗

arXiv2025

Leveraging LLMs for Scalable Non-intrusive Speech Quality Assessment

Non-intrusive speech quality assessment (SQA) systems suffer from limited training data and costly human annotations, hindering their generalization to real-time conferencing calls. In this work, we propose leveraging large language models (LLMs) as pseudo-raters for speech quality to address these data bottlenecks. We construct LibriAugmented, a dataset consisting of 101,129 speech clips with simulated degradations labeled by a fine-tuned auditory LLM (Vicuna-7b-v1.5). We compare three training strategies: using human-labeled data, using LLM-labeled data, and a two-stage approach (pretraining on LLM labels, then fine-tuning on human labels), using both DNSMOS Pro and DeePMOS. We test on several datasets across languages and quality degradations. While LLM-labeled training yields mixed results compared to human-labeled training, we provide empirical evidence that the two-stage approach improves the generalization performance (e.g., DNSMOS Pro achieves 0.63 vs. 0.55 PCC on NISQA_TEST_LIVETALK and 0.73 vs. 0.65 PCC on Tencent with reverb). Our findings demonstrate the potential of using LLMs as scalable pseudo-raters for speech quality assessment, offering a cost-effective solution to the data limitation problem.

eess.AS↗

arXiv2025

Multivariate Probabilistic Assessment of Speech Quality

The mean opinion score (MOS) is a standard metric for assessing speech quality, but its singular focus fails to identify specific distortions when low scores are observed. The NISQA dataset addresses this limitation by providing ratings across four additional dimensions: noisiness, coloration, discontinuity, and loudness, alongside MOS. In this paper, we extend the explored univariate MOS estimation to a multivariate framework by modeling these dimensions jointly using a multivariate Gaussian distribution. Our approach utilizes Cholesky decomposition to predict covariances without imposing restrictive assumptions and extends probabilistic affine transformations to a multivariate context. Experimental results show that our model performs on par with state-of-the-art methods in point estimation, while uniquely providing uncertainty and correlation estimates across speech quality dimensions. This enables better diagnosis of poor speech quality and informs targeted improvements.

eess.AS↗

arXiv2025

Impairments are Clustered in Latents of Deep Neural Network-based Speech Quality Models

In this article, we provide an experimental observation: Deep neural network (DNN) based speech quality assessment (SQA) models have inherent latent representations where many types of impairments are clustered. While DNN-based SQA models are not trained for impairment classification, our experiments show good impairment classification results in an appropriate SQA latent representation. We investigate the clustering of impairments using various kinds of audio degradations that include different types of noises, waveform clipping, gain transition, pitch shift, compression, reverberation, etc. To visualize the clusters we perform classification of impairments in the SQA-latent representation domain using a standard k-nearest neighbor (kNN) classifier. We also develop a new DNN-based SQA model, named DNSMOS+, to examine whether an improvement in SQA leads to an improvement in impairment classification. The classification accuracy is 94% for LibriAugmented dataset with 16 types of impairments and 54% for ESC-50 dataset with 50 types of real noises.

eess.AS↗

arXiv2025

Connections on a principal Lie groupoid bundle and representations up to homotopy

A Lie groupoid principal $\mbbX$ bundle is a surjective submersion $\pi\colon P\to M$ with an action of $\mathbb{X}$ on $P$ with certain additional conditions. This paper offers a suitable definition for the notion of a connection on such bundles. Although every Lie groupoid $\mathbb{X}$ has its associated Lie algebroid $A:=1^*\ker ds\to X_0$, it does not admit a natural action on its Lie algebroid. There is no natural action of $\mathbb{X}$ on $TP$ either. Choosing a connection $\mathbb{H}\subset TX_1$ on the Lie groupoid $\mathbb{X},$ and considering its induced action up to homotopy of $\mathbb{X}$ on graded vector bundle $TX_0\oplus A,$ we prove the existence of a short exact sequence of diffeological groupoids over the discrete category $M$ (with appropriate vector space structures on the fibres) for the $\mbbX$ bundle $\pi\colon P\to M.$ We introduce a notion of connection on $\mbbX$ bundle $\pi\colon P\to M,$ and show that such a connection $\omega$ splits the sequence. Finally, we show that a connection pair $(\omega, \mathbb{H})$ on $\mbbX$ bundle $\pi\colon P\to M$ is isomorphic to any other connection pair.}

math.DG↗

arXiv2024

AI-Aided Kalman Filters

The Kalman filter (KF) and its variants are among the most celebrated algorithms in signal processing. These methods are used for state estimation of dynamic systems by relying on mathematical representations in the form of simple state-space (SS) models, which may be crude and inaccurate descriptions of the underlying dynamics. Emerging data-centric artificial intelligence (AI) techniques tackle these tasks using deep neural networks (DNNs), which are model-agnostic. Recent developments illustrate the possibility of fusing DNNs with classic Kalman-type filtering, obtaining systems that learn to track in partially known dynamics. This article provides a tutorial-style overview of design approaches for incorporating AI in aiding KF-type algorithms. We review both generic and dedicated DNN architectures suitable for state estimation, and provide a systematic presentation of techniques for fusing AI tools with KFs and for leveraging partial SS modeling and data, categorizing design approaches into task-oriented and SS model-oriented. The usefulness of each approach in preserving the individual strengths of model-based KFs and data-driven DNNs is investigated in a qualitative and quantitative study, whose code is publicly available, illustrating the gains of hybrid model-based/data-driven designs. We also discuss existing challenges and future research directions that arise from fusing AI and Kalman-type algorithms.

cs.LG↗

arXiv2024

Near-Field ISAC in 6G: Addressing Phase Nonlinearity via Lifted Super-Resolution

Integrated sensing and communications (ISAC) is a promising component of 6G networks, fusing communication and radar technologies to facilitate new services. Additionally, the use of extremely large-scale antenna arrays (ELAA) at the ISAC common receiver not only facilitates terahertz-rate communication links but also significantly enhances the accuracy of target detection in radar applications. In practical scenarios, communication scatterers and radar targets often reside in close proximity to the ISAC receiver. This, combined with the use of ELAA, fundamentally alters the electromagnetic characteristics of wireless and radar channels, shifting from far-field planar-wave propagation to near-field spherical wave propagation. Under the far-field planar-wave model, the phase of the array response vector varies linearly with the antenna index. In contrast, in the near-field spherical wave model, this phase relationship becomes nonlinear. This shift presents a fundamental challenge: the widely-used Fourier analysis can no longer be directly applied for target detection and communication channel estimation at the ISAC common receiver. In this work, we propose a feasible solution to address this fundamental issue. Specifically, we demonstrate that there exists a high-dimensional space in which the phase nonlinearity can be expressed as linear. Leveraging this insight, we develop a lifted super-resolution framework that simultaneously performs communication channel estimation and extracts target parameters with high precision.

eess.SP↗

arXiv2024

Semi-Supervised Model-Free Bayesian State Estimation from Compressed Measurements

We consider data-driven Bayesian state estimation from compressed measurements (BSCM) of a model-free process. The dimension of the temporal measurement vector is lower than that of the temporal state vector to be estimated, leading to an under-determined inverse problem. The underlying dynamical model of the state's evolution is unknown for a `model-free process.' Hence, it is difficult to use traditional model-driven methods, for example, Kalman and particle filters. Instead, we consider data-driven methods. We experimentally show that two existing unsupervised learning-based data-driven methods fail to address the BSCM problem in a model-free process. The methods are -- data-driven nonlinear state estimation (DANSE) and deep Markov model (DMM). While DANSE provides good predictive/forecasting performance to model the temporal measurement data as a time series, its unsupervised learning lacks suitable regularization for tackling the BSCM task. We then propose a semi-supervised learning approach and develop a semi-supervised learning-based DANSE method, referred to as SemiDANSE. In SemiDANSE, we use a large amount of unlabelled data along with a limited amount of labelled data, i.e., pairwise measurement-and-state data, which provides the desired regularization. Using {benchmark chaotic dynamical systems}, we {empirically} show that the data-driven SemiDANSE provides competitive state estimation performance for BSCM {using a handful of different measurement systems}, against a hybrid method called KalmanNet and two model-driven methods (extended Kalman filter and unscented Kalman filter) that know the dynamical models exactly.

eess.SP↗

arXiv2023

Compressed Sensing of Generative Sparse-latent (GSL) Signals

We consider reconstruction of an ambient signal in a compressed sensing (CS) setup where the ambient signal has a neural network based generative model. The generative model has a sparse-latent input and we refer to the generated ambient signal as generative sparse-latent signal (GSL). The proposed sparsity inducing reconstruction algorithm is inherently non-convex, and we show that a gradient based search provides a good reconstruction performance. We evaluate our proposed algorithm using simulated data.

cs.LG↗

arXiv2023

Connections on Lie groupoids and Chern-Weil theory

Let $\mathbb{X}=[X_1\rightrightarrows X_0]$ be a Lie groupoid equipped with a connection, given by a smooth distribution $\mathcal{H} \subset T X_1$ transversal to the fibers of the source map. Under the assumption that the distribution $\mathcal{H}$is integrable, we define a version of de Rham cohomology for the pair $(\mathbb{X}, \mathcal{H})$, and we study connections on principal $G$-bundles over $(\mathbb{X}, \mathcal{H})$ in terms of the associated Atiyah sequence of vector bundles. We also discuss associated constructions for differentiable stacks. Finally, we develop the corresponding Chern-Weil theory and describe characteristic classes of principal G-bundles over a pair $(\mathbb{X}, \mathcal{H})$.

math.DG↗

arXiv2023

Identifying and analysing toxic actors and communities on Facebook by employing network analysis

There has been an increasingly widespread agreement among both academic circles and the general public that the Social Media Platforms (SMPs) play a central role in the dissemination of harmful and negative sentiment content in a coordinated manner. A substantial body of recent scholarly research has demonstrated the ways in which hateful content, political propaganda, and targeted messaging on SMPs have contributed to serious real-world consequences. Adopting inspirations from graph theory, in this paper we apply novel network and community finding algorithms over a representative Facebook dataset (n=608,417) which we have scrapped through 630 pages. By applying Girvan-Newman algorithm over the historical dataset our analysis finds five communities of coordinated networks of actors, within the contexts of Indian far-right Hindutva discourse. This work further paves the path for future potentials of applying such novel network analysis algorithms to SMPs, in order to automatically identify toxic coordinated communities and sub-communities, and to possibly resist real-world threats emerging from information dissemination in the SMPs.

cs.SI↗