arXiv subjects

Kyogu Lee

Publications and source records attributed to Kyogu Lee.

At least 19 recordsLinked to original sources

arXiv2026

Auditory Illusion Benchmark for Large Audio Language Models

Perceptual illusions have long served as crucial probes into human cognition, revealing biases and limitations of perception. In the auditory domain, such illusions provide a unique lens for testing whether Large Audio Language Models (LALMs) replicate human perceptual tendencies. Despite their importance, most benchmarks focus on visual illusions or general audio tasks, leaving auditory illusions underexplored. To this end, we present AIB, the first auditory illusion benchmark for LALMs, covering ten representative illusions across music, sound, and speech, each annotated for the presence of knowledge-based priors. Our methodology pairs model evaluation with controlled human listening studies, enabling direct comparison of responses. Results show systematic differences: while most LALMs remain signal-faithful on low-level acoustic illusions, several exhibit more human-like responses when linguistic or musical priors are involved, although no model matches the human perceptual profile. These findings highlight the current limitations of LALMs as cognitive models. By establishing auditory illusions as a rigorous testbed, our work offers a new perspective for probing neural black-box models and advancing understanding of auditory cognition. AIB is publicly available at https://github.com/gillosae/aib.

cs.SD↗

arXiv2026

Vocal Music under Phoneme-Conditional Analysis

The vocal music of each language carries a distinctive sonic identity, even without instrumental accompaniment. We ask whether these differences are measurable and traceable to specific phonemes. To tackle this question, we introduce phoneme-conditional analysis, which isolates the acoustic effect of typologically distinctive phonemes by comparing marker syllables against matched non-marker controls within the same song, holding singer, melody, and genre constant. Across nine typologically diverse languages and thousands of songs, we measure effects along five acoustic dimensions. Song-level profiles built from these effects identify the language of an unaccompanied vocal at 85.5% balanced accuracy in a nine-way classification with folds grouped by artist; whether the separability arises by accumulation of the phoneme-local effects themselves is left open. Our findings suggest that phonological structure leaves systematic and measurable traces in how each language is sung.

cs.SD↗

arXiv2026

LipCoder: Voice-Enabled Coding Toolkit

AI-assisted programming environments have accelerated software development, giving rise to new paradigms like vibe coding. However, their benefits remain largely inaccessible to visually impaired programmers, as existing screen readers and assistive tools offer limited support for these emerging workflows. We introduce LipCoder, a voice-centric programming toolkit designed to deliver editor-level functionality through auditory and speech-based interfaces. LipCoder offers features comprising speech feedback and earcon cues for comprehension and validation, as well as natural language input for navigation and modification. In an exploratory evaluation, 5 visually impaired programmers performed a series of coding tasks comparing LipCoder with a baseline of VSCode, Copilot, and VoiceOver. Quantitative trends and qualitative feedback point to directions for auditory-first design that may broaden accessibility in speech-driven coding environments.

cs.HC↗

arXiv2026

Exploring the Design Space of Representation Learning for Audio Transformations

Neural audio representation learning has enabled a range of content-oriented applications, but the resulting features remain limited for tasks involving audio processing. Furthermore, it is not obvious what processing-aware representations should capture: the processing itself, abstracted away from source content, or the processed audio that retains it. Existing approaches implicitly commit to one or the other and also differ in their models, data, and evaluation, obscuring which design choices drive their behavior. We address both questions within a unified framework of three objectives: processing consistency, description alignment, and equivariance via forward prediction. We compare all combinations of the objectives under a controlled setup and reveal their relative strengths and interactions. Our framework produces both a transformation embedding and a processed-audio embedding, and we find that the two play complementary roles: distance-based tasks favor the former, while probe-based tasks favor the latter. Combined with improvements in network architecture and training pipeline, our representations outperform prior baselines across retrieval, probe-based evaluation, and style transfer.

cs.SD↗

arXiv2026

Breaking Shortcut Learning for Cross-Trial EEG-Guided Target Speech Extraction via Two-Stage Training

Recent end-to-end models for EEG-guided target speech extraction report impressive results, underscoring potential for neuro-steered hearing technologies. However, our analysis reveals that high within-trial performance can be driven by trial-specific EEG structure that acts as shortcuts for target selection, leading to poor generalization on unseen trials. To overcome this gap, we propose TRUST-TSE, a two-stage framework to mitigate shortcut learning. By introducing contrastive pretraining with attended-speaker negative sampling, we encourage the EEG encoder to capture fine-grained EEG--speech alignment while suppressing trial-identity cues. We also employ a confidence-weighted extraction objective based on EEG--source similarity to guide extraction using the learned representations. Experiments on KUL and DTU datasets show that TRUST-TSE outperforms end-to-end baselines under strict cross-trial protocols, addressing a key reliability bottleneck of existing approaches.

eess.AS↗

arXiv2026

Toward Open-Set Speaker Attribute Prediction with Keyword-Appended LLM Embeddings

Understanding speaker attributes is crucial for voice-related applications, yet conventional approaches rely on fixed categorical labels, lacking semantic richness and zero-shot generalizability. We propose a novel framework for open-set speaker attribute prediction leveraging Large Language Model (LLM) embeddings to represent attributes in a continuous semantic space. To bridge the cross-modal gap, we introduce a keyword-appending strategy that structures broad semantic representations into a compact, discriminative manifold. Furthermore, we employ a top-k negative loss to establish robust decision boundaries in crowded semantic regions. Experimental results on LibriTTS-P demonstrate that our method outperforms closed-set benchmarks and generalizes effectively to unseen synonyms. Geometric analysis suggests that our strategies regularize the embedding manifold, balancing semantic cohesion with predictive clarity.

cs.SD↗

arXiv2026

Fast Speech Foundation Model Distillation Using Interleaved Stacking

Distilling a large speech foundation model (SFM) into an efficient student model has been successfully applied to low-resource environments. Although distillation reduces inference latency, it requires an additional student model training. However, the training efficiency of SFM distillation remains underexplored. In this work, we explore training acceleration of SFM distillation to speed up model deployment. We examine the potential of stacking, in which the model depth is progressively increased through training until the target model depth is reached. While existing stacking methods improve training speed, they suffer from performance degradation. To handle this limitation, we propose interleaved stacking, a novel stacking method that consistently preserves layer position throughout the stacking process. This property is particularly critical in SFMs, in which each layer encodes distinct layer-specific knowledge. We validate the effectiveness of the proposed method on SUPERB.

eess.AS↗

arXiv2026

Differentiable Acoustic Radiance Transfer

Geometric acoustics is an efficient framework for room acoustics modeling, governed by the canonical time-dependent rendering equation. Acoustic radiance transfer (ART) solves the equation by discretization, modeling time- and direction-dependent energy exchange between surface patches with flexible material properties. We introduce DART, an efficient, differentiable implementation of ART that enables gradient-based optimization of material properties. We evaluate DART on a simpler variant of acoustic field learning that aims to predict energy responses for novel source-receiver configurations. Experimental results demonstrate that DART generalizes better under sparse measurement scenarios than existing signal processing and neural network baselines, while maintaining simplicity and full interpretability. We open-source our implementation.

cs.SD↗

arXiv2026

Cross-Modal Bottleneck Fusion For Noise Robust Audio-Visual Speech Recognition

Audio-Visual Speech Recognition (AVSR) leverages both acoustic and visual cues to improve speech recognition under noisy conditions. A central question is how to design a fusion mechanism that allows the model to effectively exploit visual information when the audio signal is degraded, while maintaining strong performance on clean speech. We propose CoBRA (Cross-modal Bottleneck for Robust AVSR), a bottleneck-based fusion framework that introduces a compact set of learnable tokens to mediate cross-modal exchange. By regulating information flow through these tokens, the audio stream can reliably access essential visual cues even under adverse or out-of-domain noise. Despite limited training data, our model surpasses comparable baselines and remains competitive with large-scale systems through noise-adaptive fusion, demonstrating both efficiency and robustness. Ablation studies highlight that the depth of fusion is the most critical factor, underscoring its importance in designing robust AVSR systems.

eess.AS↗

arXiv2026

EMG-to-Speech with Fewer Channels

Surface electromyography (EMG) is a promising modality for silent speech interfaces, but its effectiveness depends heavily on sensor placement and channel availability. In this work, we investigate the contribution of individual and combined EMG channels to speech reconstruction performance. Our findings reveal that while certain EMG channels are individually more informative, the highest performance arises from subsets that leverage complementary relationships among channels. We also analyzed phoneme classification accuracy under channel ablations and observed interpretable patterns reflecting the anatomical roles of the underlying muscles. To address performance degradation from channel reduction, we pretrained models on full 8-channel data using random channel dropout and fine-tuned them on reduced-channel subsets. Fine-tuning consistently outperformed training from scratch for 4 - 6 channel settings, with the best dropout strategy depending on the number of channels. These results suggest that performance degradation from sensor reduction can be mitigated through pretraining and channel-aware design, supporting the development of lightweight and practical EMG-based silent speech systems.

cs.SD↗

arXiv2026

Speaking Without Sound: Multi-speaker Silent Speech Voicing with Facial Inputs Only

In this paper, we introduce a novel framework for generating multi-speaker speech without relying on any audible inputs. Our approach leverages silent electromyography (EMG) signals to capture linguistic content, while facial images are used to match with the vocal identity of the target speaker. Notably, we present a pitch-disentangled content embedding that enhances the extraction of linguistic content from EMG signals. Extensive analysis demonstrates that our method can generate multi-speaker speech without any audible inputs and confirms the effectiveness of the proposed pitch-disentanglement approach.

cs.SD↗

arXiv2026

LipSody: Lip-to-Speech Synthesis with Enhanced Prosody Consistency

Lip-to-speech synthesis aims to generate speech audio directly from silent facial video by reconstructing linguistic content from lip movements, providing valuable applications in situations where audio signals are unavailable or degraded. While recent diffusion-based models such as LipVoicer have demonstrated impressive performance in reconstructing linguistic content, they often lack prosodic consistency. In this work, we propose LipSody, a lip-to-speech framework enhanced for prosody consistency. LipSody introduces a prosody-guiding strategy that leverages three complementary cues: speaker identity extracted from facial images, linguistic content derived from lip movements, and emotional context inferred from face video. Experimental results demonstrate that LipSody substantially improves prosody-related metrics, including global and local pitch deviations, energy consistency, and speaker similarity, compared to prior approaches.

cs.SD↗

arXiv2026

Rethinking Speech Representation Aggregation in Speech Enhancement: A Phonetic Mutual Information Perspective

Recent speech enhancement (SE) models increasingly leverage self-supervised learning (SSL) representations for their rich semantic information. Typically, intermediate features are aggregated into a single representation via a lightweight adaptation module. However, most SSL models are not trained for noise robustness, which can lead to corrupted semantic representations. Moreover, the adaptation module is trained jointly with the SE model, potentially prioritizing acoustic details over semantic information, contradicting the original purpose. To address this issue, we first analyze the behavior of SSL models on noisy speech from an information-theoretic perspective. Specifically, we measure the mutual information (MI) between the corrupted SSL representations and the corresponding phoneme labels, focusing on preservation of linguistic contents. Building upon this analysis, we introduce the linguistic aggregation layer, which is pre-trained to maximize MI with phoneme labels (with optional dynamic aggregation) and then frozen during SE training. Experiments show that this decoupled approach improves Word Error Rate (WER) over jointly optimized baselines, demonstrating the benefit of explicitly aligning the adaptation module with linguistic contents.

cs.SD↗

arXiv2025

MGE-LDM: Joint Latent Diffusion for Simultaneous Music Generation and Source Extraction

We present MGE-LDM, a unified latent diffusion framework for simultaneous music generation, source imputation, and query-driven source separation. Unlike prior approaches constrained to fixed instrument classes, MGE-LDM learns a joint distribution over full mixtures, submixtures, and individual stems within a single compact latent diffusion model. At inference, MGE-LDM enables (1) complete mixture generation, (2) partial generation (i.e., source imputation), and (3) text-conditioned extraction of arbitrary sources. By formulating both separation and imputation as conditional inpainting tasks in the latent space, our approach supports flexible, class-agnostic manipulation of arbitrary instrument sources. Notably, MGE-LDM can be trained jointly across heterogeneous multi-track datasets (e.g., Slakh2100, MUSDB18, MoisesDB) without relying on predefined instrument categories. Audio samples are available at our project page: https://yoongi43.github.io/MGELDM_Samples/.

cs.SD↗

arXiv2025

Reverse Engineering of Music Mixing Graphs with Differentiable Processors and Iterative Pruning

Reverse engineering of music mixes aims to uncover how dry source signals are processed and combined to produce a final mix. We extend the prior works to reflect the compositional nature of mixing and search for a graph of audio processors. First, we construct a mixing console, applying all available processors to every track and subgroup. With differentiable processor implementations, we optimize their parameters with gradient descent. Then, we repeat the process of removing negligible processors and fine-tuning the remaining ones. This way, the quality of the full mixing console can be preserved while removing approximately two-thirds of the processors. The proposed method can be used not only to analyze individual music mixes but also to collect large-scale graph data that can be used for downstream tasks, e.g., automatic mixing. Especially for the latter purpose, efficient implementation of the search is crucial. To this end, we present an efficient batch-processing method that computes multiple processors in parallel. We also exploit the "dry/wet" parameter of the processors to accelerate the search. Extensive quantitative and qualitative analyses are conducted to evaluate the proposed method's performance, behavior, and computational cost.

cs.SD↗

arXiv2025

Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion

Flow and diffusion models have demonstrated strong performance and training stability across various tasks but lack two critical properties of simulation-based methods: freedom of dimensionality and adaptability to different inference trajectories. To address this limitation, we propose the Multidimensional Adaptive Coefficient (MAC), a plug-in module for flow and diffusion models that extends conventional unidimensional coefficients to multidimensional ones and enables inference trajectory-wise adaptation. MAC is trained via simulation-based feedback through adversarial refinement. Empirical results across diverse frameworks and datasets demonstrate that MAC enhances generative quality with high training efficiency. Consequently, our work offers a new perspective on inference trajectory optimality, encouraging future research to move beyond vector field design and to leverage training-efficient, simulation-based optimization.

cs.LG↗

arXiv2025

Vo-Ve: An Explainable Voice-Vector for Speaker Identity Evaluation

In this paper, we propose Vo-Ve, a novel voice-vector embedding that captures speaker identity. Unlike conventional speaker embeddings, Vo-Ve is explainable, as it contains the probabilities of explicit voice attribute classes. Through extensive analysis, we demonstrate that Vo-Ve not only evaluates speaker similarity competitively with conventional techniques but also provides an interpretable explanation in terms of voice attributes. We strongly believe that Vo-Ve can enhance evaluation schemes across various speech tasks due to its high-level explainability.

cs.SD↗

arXiv2025

Song Form-aware Full-Song Text-to-Lyrics Generation with Multi-Level Granularity Syllable Count Control

Lyrics generation presents unique challenges, particularly in achieving precise syllable control while adhering to song form structures such as verses and choruses. Conventional line-by-line approaches often lead to unnatural phrasing, underscoring the need for more granular syllable management. We propose a framework for lyrics generation that enables multi-level syllable control at the word, phrase, line, and paragraph levels, aware of song form. Our approach generates complete lyrics conditioned on input text and song form, ensuring alignment with specified syllable constraints. Generated lyrics samples are available at: https://tinyurl.com/lyrics9999

cs.CL↗