Trending in Electrical Engineering and Systems Science
ReStyle-TTS: Relative and Continuous Style Control for Zero-Shot Speech Synthesis
Zero-shot text-to-speech models can clone a speaker's timbre from a short reference audio, but they also strongly inherit the speaking style present in the reference. As a result, synthesizing speech with a desired style often requires carefully selecting reference audio, which is impractical when only limited or mismatched references are available. While recent controllable TTS methods attempt to address this issue, they typically rely on absolute style targets and discrete textual prompts, and therefore do not support continuous and reference-relative style control. We propose ReStyle-TTS, a framework that enables continuous and reference-relative style control in zero-shot TTS. Our key insight is that effective style control requires first reducing the model's implicit dependence on reference style before introducing explicit control mechanisms. To this end, we introduce Decoupled Classifier-Free Guidance (DCFG), which independently controls text and reference guidance, reducing reliance on reference style while preserving text fidelity. On top of this, we apply style-specific LoRAs together with Orthogonal LoRA Fusion to enable continuous and disentangled multi-attribute control, and introduce a Timbre Consistency Optimization module to mitigate timbre drift caused by weakened reference guidance. Experiments show that ReStyle-TTS enables user-friendly, continuous, and relative control over pitch, energy, and multiple emotions while maintaining intelligibility and speaker timbre, and performs robustly in challenging mismatched reference-target style scenarios.
2601.03632
Jan 2026Audio and Speech Processing
Towards Fine-Grained and Multi-Granular Contrastive Language-Speech Pre-training
Modeling fine-grained speaking styles remains challenging for language-speech representation pre-training, as existing speech-text models are typically trained with coarse captions or task-specific supervision, and scalable fine-grained style annotations are unavailable. We present FCaps, a large-scale dataset with fine-grained free-text style descriptions, encompassing 47k hours of speech and 19M fine-grained captions annotated via a novel end-to-end pipeline that directly grounds detailed captions in audio, thereby avoiding the error propagation caused by LLM-based rewriting in existing cascaded pipelines. Evaluations using LLM-as-a-judge demonstrate that our annotations surpass existing cascaded annotations in terms of correctness, coverage, and naturalness. Building on FCaps, we propose CLSP, a contrastive language-speech pre-trained model that integrates global and fine-grained supervision, enabling unified representations across multiple granularities. Extensive experiments demonstrate that CLSP learns fine-grained and multi-granular speech-text representations that perform reliably across global and fine-grained speech-text retrieval, zero-shot paralinguistic classification, and speech style similarity scoring, with strong alignment to human judgments. All resources will be made publicly available.
2601.03065
Jan 2026Audio and Speech Processing
The Dawn of Agentic EDA: A Survey of Autonomous Digital Chip Design
This survey provides a comprehensive overview of the integration of Generative AI and Agentic AI within the field of Digital Electronic Design Automation (EDA). The paper first reviews the paradigmatic evolution from traditional Computer-Aided Design (CAD) to AI-assisted EDA (AI4EDA), and finally to the emerging AI-Native and Agentic design paradigms. We detail the application of these paradigms across the digital chip design flow, including the construction of agentic cognitive architectures based on multimodal foundation models, frontend RTL code generation and intelligent verification, and backend physical design featuring algorithmic innovations and tool orchestration. We validate these methodologies through integrated case studies, demonstrating practical viability from microarchitecture definition to GDSII. Special emphasis is placed on the potential for cross-stage feedback loops where agents utilize backend PPA metrics to autonomously refine frontend logic. Furthermore, this survey delves into the dual-faceted impact on security, covering novel adversarial risks, automated vulnerability repair, and privacy-preserving infrastructure. Finally, the paper critically summarizes current challenges related to hallucinations, data scarcity, and black-box tools, and outlines future trends towards L4 autonomous chip design. Ultimately, this work aims to define the emerging field of Agentic EDA and provide a strategic roadmap for the transition from AI-assisted tools to fully autonomous design engineers.
2512.23189
Dec 2025Systems and Control
Flow2GAN: Hybrid Flow Matching and GAN with Multi-Resolution Network for Few-step High-Fidelity Audio Generation
Existing dominant methods for audio generation include Generative Adversarial Networks (GANs) and diffusion-based methods like Flow Matching. GANs suffer from slow convergence and potential mode collapse during training, while diffusion methods require multi-step inference that introduces considerable computational overhead. In this work, we introduce Flow2GAN, a two-stage framework that combines Flow Matching training for learning generative capabilities with GAN fine-tuning for efficient few-step inference. Specifically, given audio's unique properties, we first improve Flow Matching for audio modeling through: 1) reformulating the objective as endpoint estimation, avoiding velocity estimation difficulties when involving empty regions; 2) applying spectral energy-based loss scaling to emphasize perceptually salient quieter regions. Building on these Flow Matching adaptations, we demonstrate that a further stage of lightweight GAN fine-tuning enables us to obtain one-step generator that produces high-quality audio. In addition, we develop a multi-branch network architecture that processes Fourier coefficients at different time-frequency resolutions, which improves the modeling capabilities compared to prior single-resolution designs. Experimental results indicate that our Flow2GAN delivers high-fidelity audio generation from Mel-spectrograms or discrete audio tokens, achieving better quality-efficiency trade-offs than existing state-of-the-art GAN-based and Flow Matching-based methods. Online demo samples are available at https://flow2gan.github.io, and the source code is released at https://github.com/k2-fsa/Flow2GAN.
2512.23278
Dec 2025Audio and Speech Processing
QuarkAudio Technical Report
Many existing audio processing and generation models rely on task-specific architectures, resulting in fragmented development efforts and limited extensibility. It is therefore promising to design a unified framework capable of handling multiple tasks, while providing robust instruction and audio understanding and high-quality audio generation. This requires a compatible paradigm design, a powerful backbone, and a high-fidelity audio reconstruction module. To meet these requirements, this technical report introduces QuarkAudio, a decoder-only autoregressive (AR) LM-based generative framework that unifies multiple tasks. The framework includes a unified discrete audio tokenizer, H-Codec, which incorporates self-supervised learning (SSL) representations into the tokenization and reconstruction process. We further propose several improvements to H-Codec, such as a dynamic frame-rate mechanism and extending the audio sampling rate to 48 kHz. QuarkAudio unifies tasks by using task-specific conditional information as the conditioning sequence of the decoder-only LM, and predicting discrete target audio tokens in an AR manner. The framework supports a wide range of audio processing and generation tasks, including speech restoration (SR), target speaker extraction (TSE), speech separation (SS), voice conversion (VC), and language-queried audio source separation (LASS). In addition, we extend downstream tasks to universal free-form audio editing guided by natural language instructions (including speech semantic editing and audio event editing). Experimental results show that H-Codec achieves high-quality audio reconstruction with a low frame rate, improving both the efficiency and performance of downstream audio generation, and that QuarkAudio delivers competitive or comparable performance to state-of-the-art task-specific or multi-task systems across multiple tasks.
2512.20151
Dec 2025Audio and Speech Processing
Intertemporal Hedging Demand under Epstein-Zin Preferences in a Multi-Asset Long-Run Risk Model: Evidence from Projected Pontryagin-Guided Deep Policy Optimization
I study intertemporal hedging demand in a continuous-time multi-asset long-run risk (LRR) model under Epstein--Zin (EZ) recursive preferences. The investor trades a risk-free asset and several risky assets whose drifts and volatilities depend on an Ornstein--Uhlenbeck type LRR factor. Preferences are described by EZ utility with risk aversion $R$, elasticity of intertemporal substitution $ψ$, and discount rate $δ$, so that the standard time-additive CRRA case appears as a limiting benchmark. To handle the high-dimensional consumption--investment problem, I use a projected Pontryagin-guided deep policy optimization (P-PGDPO) scheme adapted to EZ preferences. The method starts from the continuous-time Hamiltonian implied by the Pontryagin maximum principle, represents the value and costate processes with neural networks, and updates the policy along the Hamiltonian gradient. Portfolio constraints and a lower bound on wealth are enforced by explicit projection operators rather than by adding ad hoc penalties. Three main findings emerge from numerical experiments in a five-asset LRR economy: \textbf{(1)} the P-PGDPO algorithm achieves stable convergence across multiple random seeds, validating its reliability for solving high-dimensional EZ problems; \textbf{(2)} wealth floors materially reduce hedging demand by limiting the investor's ability to exploit intertemporal risk-return tradeoffs; and \textbf{(3)} the learned hedging portfolios concentrate exposure in assets with high correlation to the LRR factor, confirming that EZ agents actively hedge long-run uncertainty rather than merely following myopic rules. Because EZ preferences nest time-additive CRRA in the limit $ψ\to 1/R$, I use CRRA as an explicit diagnostic benchmark and, when needed, a warm start to stabilize training in high dimensions.
2512.15175
Dec 2025Systems and Control
LUCID: Learning-Enabled Uncertainty-Aware Certification of Stochastic Dynamical Systems
Ensuring the safety of AI-enabled systems, particularly in high-stakes domains such as autonomous driving and healthcare, has become increasingly critical. Traditional formal verification tools fall short when faced with systems that embed both opaque, black-box AI components and complex stochastic dynamics. To address these challenges, we introduce LUCID (Learning-enabled Uncertainty-aware Certification of stochastIc Dynamical systems), a verification engine for certifying safety of black-box stochastic dynamical systems from a finite dataset of random state transitions. As such, LUCID is the first known tool capable of establishing quantified safety guarantees for such systems. Thanks to its modular architecture and extensive documentation, LUCID is designed for easy extensibility. LUCID employs a data-driven methodology rooted in control barrier certificates, which are learned directly from system transition data, to ensure formal safety guarantees. We use conditional mean embeddings to embed data into a reproducing kernel Hilbert space (RKHS), where an RKHS ambiguity set is constructed that can be inflated to robustify the result to out-of-distribution behavior. A key innovation within LUCID is its use of a finite Fourier kernel expansion to reformulate a semi-infinite non-convex optimization problem into a tractable linear program. The resulting spectral barrier allows us to leverage the fast Fourier transform to generate the relaxed problem efficiently, offering a scalable yet distributionally robust framework for verifying safety. LUCID thus offers a robust and efficient verification framework, able to handle the complexities of modern black-box systems while providing formal guarantees of safety. These unique capabilities are demonstrated on challenging benchmarks.
2512.11750
Dec 2025Systems and Control
2512.11734Model Error Resonance: The Geometric Nature of Error Dynamics
This paper introduces a geometric theory of model error, treating true and model dynamics as geodesic flows generated by distinct affine connections on a smooth manifold. When these connections differ, the resulting trajectory discrepancy--termed the Latent Error Dynamic Response (LEDR)--acquires an intrinsic dynamical structure governed by curvature. We show that the LEDR satisfies a Jacobi-type equation, where curvature mismatch acts as an explicit forcing term. In the important case of a flat model connection, the LEDR reduces to a classical Jacobi field on the true manifold, causing Model Error Resonance (MER) to emerge under positive sectional curvature. The theory is extended to a discrete-time analogue, establishing that this geometric structure and its resonant behavior persist in sampled systems. A closed-form analysis of a sphere--plane example demonstrates that curvature can be inferred directly from the LEDR evolution. This framework provides a unified geometric interpretation of structured error dynamics and offers foundational tools for curvature-informed model validation.
2512.11734
Dec 2025Systems and Control
A Universal Harmonic Discriminator for High-quality GAN-based Vocoder
With the emergence of GAN-based vocoders, the discriminator, as a crucial component, has been developed recently. In our work, we focus on improving the time-frequency based discriminator. Particularly, Short-Time Fourier Transform (STFT) representation is usually used as input of time-frequency based discriminator. However, the STFT spectrogram has the same frequency resolution at different frequency bins, which results in an inferior performance, especially for singing voices. Motivated by this, we propose a universal harmonic discriminator for dynamic frequency resolution modeling and harmonic tracking. Specifically, we design a harmonic filter with learnable triangular band-pass filter banks, where each frequency bin has a flexible bandwidth. Additionally, we add a half-harmonic to capture fine-grained harmonic relationships at low-frequency band. Experiments on speech and singing datasets validate the effectiveness of the proposed discriminator on both subjective and objective metrics.
2512.03486
Dec 2025Audio and Speech Processing
ECO: Energy-Constrained Operator Learning for Chaotic Dynamics with Boundedness Guarantees
Chaos is a fundamental feature of many complex dynamical systems, including weather systems and fluid turbulence. These systems are inherently difficult to predict due to their extreme sensitivity to initial conditions. Many chaotic systems are dissipative and ergodic, motivating data-driven models that aim to learn invariant statistical properties over long time horizons. While recent models have shown empirical success in preserving invariant statistics, they are prone to generating unbounded predictions, which prevent meaningful statistics evaluation. To overcome this, we introduce the Energy-Constrained Operator (ECO) that simultaneously learns the system dynamics while enforcing boundedness in predictions. We leverage concepts from control theory to develop algebraic conditions based on a learnable energy function, ensuring the learned dynamics is dissipative. ECO enforces these algebraic conditions through an efficient closed-form quadratic projection layer, which provides provable trajectory boundedness. To our knowledge, this is the first work establishing such formal guarantees for data-driven chaotic dynamics models. Additionally, the learned invariant level set provides an outer estimate for the strange attractor, a complex structure that is computationally intractable to characterize. We demonstrate empirical success in ECO's ability to generate stable long-horizon forecasts, capturing invariant statistics on systems governed by chaotic PDEs, including the Kuramoto--Sivashinsky and the Navier--Stokes equations.
2512.01984
Dec 2025Systems and Control
AI-Driven Optimization under Uncertainty for Mineral Processing Operations
The global capacity for mineral processing must expand rapidly to meet the demand for critical minerals, which are essential for building the clean energy technologies necessary to mitigate climate change. However, the efficiency of mineral processing is severely limited by uncertainty, which arises from both the variability of feedstock and the complexity of process dynamics. To optimize mineral processing circuits under uncertainty, we introduce an AI-driven approach that formulates mineral processing as a Partially Observable Markov Decision Process (POMDP). We demonstrate the capabilities of this approach in handling both feedstock uncertainty and process model uncertainty to optimize the operation of a simulated, simplified flotation cell as an example. We show that by integrating the process of information gathering (i.e., uncertainty reduction) and process optimization, this approach has the potential to consistently perform better than traditional approaches at maximizing an overall objective, such as net present value (NPV). Our methodological demonstration of this optimization-under-uncertainty approach for a synthetic case provides a mathematical and computational framework for later real-world application, with the potential to improve both the laboratory-scale design of experiments and industrial-scale operation of mineral processing circuits without any additional hardware.
2512.01977
Dec 2025Systems and Control
The Equivalence of Fast Algorithms for Convolution, Parallel FIR Filters, Polynomial Modular Multiplication, and Pointwise Multiplication in DFT/NTT Domain
Fast time-domain algorithms have been developed in signal processing applications to reduce the multiplication complexity. For example, fast convolution structures using Cook-Toom and Winograd algorithms are well understood. Short length fast convolutions can be iterated to obtain fast convolution structures for long lengths. In this paper, we show that well known fast convolution structures form the basis for design of fast algorithms in four other problem domains: fast parallel filters, fast polynomial modular multiplication, and fast pointwise multiplication in the DFT and NTT domains. Fast polynomial modular multiplication and fast pointwise multiplication problems are important for cryptosystem applications such as post-quantum cryptography and homomorphic encryption. By establishing the equivalence of these problems, we show that a fast structure from one domain can be used to design a fast structure for another domain. This understanding is important as there are many well known solutions for fast convolution that can be used in other signal processing and cryptosystem applications.
2512.01974
Dec 2025Signal Processing
Omni-AVSR: Towards Unified Multimodal Speech Recognition with Large Language Models
Large language models (LLMs) have recently achieved impressive results in speech recognition across multiple modalities, including Auditory Speech Recognition (ASR), Visual Speech Recognition (VSR), and Audio-Visual Speech Recognition (AVSR). Despite this progress, current LLM-based approaches typically address each task independently, training separate models that raise computational and deployment resource use while missing potential cross-task synergies. They also rely on fixed-rate token compression, which restricts flexibility in balancing accuracy with efficiency. These limitations highlight the need for a unified framework that can support ASR, VSR, and AVSR while enabling elastic inference. To this end, we present Omni-AVSR, a unified audio-visual LLM that combines efficient multi-granularity training with parameter-efficient adaptation. Specifically, we adapt the matryoshka representation learning paradigm to efficiently train across multiple audio and visual granularities, reducing its inherent training resource use. Furthermore, we explore three LoRA-based strategies for adapting the backbone LLM, balancing shared and task-specific specialization. Experiments on LRS2 and LRS3 show that Omni-AVSR achieves comparable or superior accuracy to state-of-the-art baselines while training a single model at substantially lower training and deployment resource use. The model also remains robust under acoustic noise, and we analyze its scaling behavior as LLM size increases, providing insights into the trade-off between performance and efficiency.
2511.07253
Nov 2025Audio and Speech Processing