Jihwan Lee

Junhyeok Lee, Xiluo He, Jihwan Lee et al.

Neural audio codecs optimized for mel-spectrogram reconstruction often fail to preserve intelligibility. While semantic encoder distillation improves encoded representations, it does not guarantee content preservation in reconstructed speech. In this work, we demonstrate that self-supervised representation reconstruction (SSRR) loss fundamentally improves codec training and performance. First, SSRR significantly accelerates convergence, enabling competitive results using only a single GPU. Second, it enhances intelligibility by reconstructing distilled self-supervised representations from codec outputs. Third, SSRR enables high intelligibility without additional lookahead in streaming Transformer-based codecs, allowing a zero-lookahead architecture for real-time deployment. As a result, our JHCodec achieves state-of-the-art performance while maintaining minimal latency and reduced training cost. We open-source the full implementation, training pipeline, and demo on Github https://github.com/jhcodec843/jhcodec.

Gus Kristiansen, Mark Sandler, Andrey Zhmoginov et al.

In modern deep learning, the models are learned by applying gradient updates using an optimizer, which transforms the updates based on various statistics. Optimizers are often hand-designed and tuning their hyperparameters is a big part of the training process. Learned optimizers have shown some initial promise, but are generally unsuccessful as a general optimization mechanism applicable to every problem. In this work we explore a different direction: instead of learning general optimizers, we instead specialize them to a specific training environment. We propose a novel optimizer technique that learns a layer-specific linear combination of update directions provided by a set of base optimizers, effectively adapting its strategy to the specific model and dataset. When evaluated on image classification tasks, this specialized optimizer significantly outperforms both traditional off-the-shelf methods such as Adam, as well as existing general learned optimizers. Moreover, it demonstrates robust generalization with respect to model initialization, evaluating on unseen datasets, and training durations beyond its meta-training horizon.

Jihwan Lee, Joun Yeop Lee, Heejin Choi et al.

Intonations play an important role in delivering the intention of a speaker. However, current end-to-end TTS systems often fail to model proper intonations. To alleviate this problem, we propose a novel, intuitive method to synthesize speech in different intonations using predefined intonation templates. Prior to TTS model training, speech data are grouped into intonation templates in an unsupervised manner. Two proposed modules are added to the end-to-end TTS framework: an intonation predictor and an intonation encoder. The intonation predictor recommends a suitable intonation template to the given text. The intonation encoder, attached to the text encoder output, synthesizes speech abiding the requested intonation template. Main contributions of our paper are: (a) an easy-to-use intonation control system covering a wide range of users; (b) better performance in wrapping speech in a requested intonation with improved objective and subjective evaluation; and (c) incorporating a pre-trained language model for intonation modelling. Audio samples are available at https://srtts.github.io/IntoTTS.

Simon Pistrosch, Kleanthis Avramidis, Tiantian Feng et al.

The expression of affect is integral to spoken communication, yet, its link to underlying articulatory execution remains unclear. Measures of articulatory muscle activity such as EMG could reveal how speech production is modulated by emotion alongside acoustic speech analyses. We investigate affect decoding from facial and neck surface electromyography (sEMG) during phonated and silent speech production. For this purpose, we introduce a dataset comprising 2,780 utterances from 12 participants across 3 tasks, on which we evaluate both intra- and inter-subject decoding using a range of features and model embeddings. Our results reveal that EMG representations reliably discriminate frustration with up to 0.845 AUC, and generalize well across articulation modes. Our ablation study further demonstrates that affective signatures are embedded in facial motor activity and persist in the absence of phonation, highlighting the potential of EMG sensing for affect-aware silent speech interfaces.

Seung hee Choi, MinJu Jeon, Hyunwoo Oh et al.

Existing retrieval-augmented approaches for Dense Video Captioning (DVC) often fail to achieve accurate temporal segmentation aligned with true event boundaries, as they rely on heuristic strategies that overlook ground truth event boundaries. The proposed framework, \textbf{STaRC}, overcomes this limitation by supervising frame-level saliency through a highlight detection module. Note that the highlight detection module is trained on binary labels derived directly from DVC ground truth annotations without the need for additional annotation. We also propose to utilize the saliency scores as a unified temporal signal that drives retrieval via saliency-guided segmentation and informs caption generation through explicit Saliency Prompts injected into the decoder. By enforcing saliency-constrained segmentation, our method produces temporally coherent segments that align closely with actual event transitions, leading to more accurate retrieval and contextually grounded caption generation. We conduct comprehensive evaluations on the YouCook2 and ViTT benchmarks, where STaRC achieves state-of-the-art performance across most of the metrics. Our code is available at https://github.com/ermitaju1/STaRC

Hosung Lee, Sejin Kim, Seungpil Lee et al.

This paper introduces ARCLE, an environment designed to facilitate reinforcement learning research on the Abstraction and Reasoning Corpus (ARC). Addressing this inductive reasoning benchmark with reinforcement learning presents these challenges: a vast action space, a hard-to-reach goal, and a variety of tasks. We demonstrate that an agent with proximal policy optimization can learn individual tasks through ARCLE. The adoption of non-factorial policies and auxiliary losses led to performance enhancements, effectively mitigating issues associated with action spaces and goal attainment. Based on these insights, we propose several research directions and motivations for using ARCLE, including MAML, GFlowNets, and World Models.

Jihwan Lee, Woochang Sim, Sejin Kim et al.

This paper demonstrates that model-based reinforcement learning (model-based RL) is a suitable approach for the task of analogical reasoning. We hypothesize that model-based RL can solve analogical reasoning tasks more efficiently through the creation of internal models. To test this, we compared DreamerV3, a model-based RL method, with Proximal Policy Optimization, a model-free RL method, on the Abstraction and Reasoning Corpus (ARC) tasks. Our results indicate that model-based RL not only outperforms model-free RL in learning and generalizing from single tasks but also shows significant advantages in reasoning across similar tasks.

Geonwoo Cho, Jaemoon Lee, Jaegyun Im et al.

Skill-based reinforcement learning (SBRL) enables rapid adaptation in environments with sparse rewards by pretraining a skill-conditioned policy. Effective skill learning requires jointly maximizing both exploration and skill diversity. However, existing methods often face challenges in simultaneously optimizing for these two conflicting objectives. In this work, we propose a new method, Adaptive Multi-objective Projection for balancing Exploration and skill Diversification (AMPED), which explicitly addresses both: during pre-training, a gradient-surgery projection balances the exploration and diversity gradients, and during fine-tuning, a skill selector exploits the learned diversity by choosing skills suited to downstream tasks. Our approach achieves performance that surpasses SBRL baselines across various benchmarks. Through an extensive ablation study, we identify the role of each component and demonstrate that each element in AMPED is contributing to performance. We further provide theoretical and empirical evidence that, with a greedy skill selector, greater skill diversity reduces fine-tuning sample complexity. These results highlight the importance of explicitly harmonizing exploration and diversity and demonstrate the effectiveness of AMPED in enabling robust and generalizable skill learning. Project Page: https://geonwoo.me/amped/

Tiantian Feng, Kevin Huang, Anfeng Xu et al.

We present Voxlect, a novel benchmark for modeling dialects and regional languages worldwide using speech foundation models. Specifically, we report comprehensive benchmark evaluations on dialects and regional language varieties in English, Arabic, Mandarin and Cantonese, Tibetan, Indic languages, Thai, Spanish, French, German, Brazilian Portuguese, and Italian. Our study used over 2 million training utterances from 30 publicly available speech corpora that are provided with dialectal information. We evaluate the performance of several widely used speech foundation models in classifying speech dialects. We assess the robustness of the dialectal models under noisy conditions and present an error analysis that highlights modeling results aligned with geographic continuity. In addition to benchmarking dialect classification, we demonstrate several downstream applications enabled by Voxlect. Specifically, we show that Voxlect can be applied to augment existing speech recognition datasets with dialect information, enabling a more detailed analysis of ASR performance across dialectal variations. Voxlect is also used as a tool to evaluate the performance of speech generation systems. Voxlect is publicly available with the license of the RAIL family at: https://github.com/tiantiaf0627/voxlect.

Jihwan Lee, Aditya Kommineni, Tiantian Feng et al.

Speech decoding from EEG signals is a challenging task, where brain activity is modeled to estimate salient characteristics of acoustic stimuli. We propose FESDE, a novel framework for Fully-End-to-end Speech Decoding from EEG signals. Our approach aims to directly reconstruct listened speech waveforms given EEG signals, where no intermediate acoustic feature processing step is required. The proposed method consists of an EEG module and a speech module along with a connector. The EEG module learns to better represent EEG signals, while the speech module generates speech waveforms from model representations. The connector learns to bridge the distributions of the latent spaces of EEG and speech. The proposed framework is both simple and efficient, by allowing single-step inference, and outperforms prior works on objective metrics. A fine-grained phoneme analysis is conducted to unveil model characteristics of speech decoding. The source code is available here: github.com/lee-jhwn/fesde.

Agung Nugraha, Heungjun Im, Jihwan Lee

High-performance concrete offers exceptional strength and durability but requires complex mix designs involving many interdependent variables and practical constraints. While data-driven methods have advanced predictive modeling for forward design, inverse design, which focuses on determining mix compositions that achieve target performance, remains limited, particularly in design situations where some mix variables are fixed by constraints and only the remaining variables must be determined. This study proposes a cooperative neural network framework for the partial inverse design of high-performance concrete. The framework combines two coupled neural network models, an imputation model that infers the undetermined variables and a surrogate model that predicts compressive strength. Through cooperative learning, the model generates valid and performance-consistent mix designs in a single forward pass while accommodating different constraint combinations without retraining. Its performance is compared with both probabilistic and generative approaches, including Bayesian inference based on a Gaussian process surrogate and autoencoder-based models. Evaluated on a benchmark dataset, the proposed model achieves stable and higher R-squared values of 0.87-0.92 and reduces mean squared error by an average of 50 percent compared with autoencoder baselines and by an average of 70 percent compared with Bayesian inference. The results demonstrate that the cooperative neural network provides an accurate, robust, and computationally efficient foundation for constraint-aware, data-driven mix proportioning in concrete engineering.

Sean Foley, Hong Nguyen, Jihwan Lee et al.

Although many previous studies have carried out multimodal learning with real-time MRI data that captures the audio-visual kinematics of the vocal tract during speech, these studies have been limited by their reliance on multi-speaker corpora. This prevents such models from learning a detailed relationship between acoustics and articulation due to considerable cross-speaker variability. In this study, we develop unimodal audio and video models as well as multimodal models for phoneme recognition using a long-form single-speaker MRI corpus, with the goal of disentangling and interpreting the contributions of each modality. Audio and multimodal models show similar performance on different phonetic manner classes but diverge on places of articulation. Interpretation of the models' latent space shows similar encoding of the phonetic space across audio and multimodal models, while the models' attention weights highlight differences in acoustic and articulatory timing for certain phonemes.

Jihwan Lee, Kevin Huang, Kleanthis Avramidis et al.

We present a model for predicting articulatory features from surface electromyography (EMG) signals during speech production. The proposed model integrates convolutional layers and a Transformer block, followed by separate predictors for articulatory features. Our approach achieves a high prediction correlation of approximately 0.9 for most articulatory features. Furthermore, we demonstrate that these predicted articulatory features can be decoded into intelligible speech waveforms. To our knowledge, this is the first method to decode speech waveforms from surface EMG via articulatory features, offering a novel approach to EMG-based speech synthesis. Additionally, we analyze the relationship between EMG electrode placement and articulatory feature predictability, providing knowledge-driven insights for optimizing EMG electrode configurations. The source code and decoded speech samples are publicly available.

Thanathai Lertpetchpun, Yoonjeong Lee, Thanapat Trachu et al.

Many spoken languages, including English, exhibit wide variation in dialects and accents, making accent control an important capability for flexible text-to-speech (TTS) models. Current TTS systems typically generate accented speech by conditioning on speaker embeddings associated with specific accents. While effective, this approach offers limited interpretability and controllability, as embeddings also encode traits such as timbre and emotion. In this study, we analyze the interaction between speaker embeddings and linguistically motivated phonological rules in accented speech synthesis. Using American and British English as a case study, we implement rules for flapping, rhoticity, and vowel correspondences. We propose the phoneme shift rate (PSR), a novel metric quantifying how strongly embeddings preserve or override rule-based transformations. Experiments show that combining rules with embeddings yields more authentic accents, while embeddings can attenuate or overwrite rules, revealing entanglement between accent and speaker identity. Our findings highlight rules as a lever for accent control and a framework for evaluating disentanglement in speech generation.

Jihwan Lee, Tiantian Feng, Aditya Kommineni et al.

Brain-computer interfaces (BCI) offer numerous human-centered application possibilities, particularly affecting people with neurological disorders. Text or speech decoding from brain activities is a relevant domain that could augment the quality of life for people with impaired speech perception. We propose a novel approach to enhance listened speech decoding from electroencephalography (EEG) signals by utilizing an auxiliary phoneme predictor that simultaneously decodes textual phoneme sequences. The proposed model architecture consists of three main parts: EEG module, speech module, and phoneme predictor. The EEG module learns to properly represent EEG signals into EEG embeddings. The speech module generates speech waveforms from the EEG embeddings. The phoneme predictor outputs the decoded phoneme sequences in text modality. Our proposed approach allows users to obtain decoded listened speech from EEG signals in both modalities (speech waveforms and textual phoneme sequences) simultaneously, eliminating the need for a concatenated sequential pipeline for each modality. The proposed approach also outperforms previous methods in both modalities. The source code and speech samples are publicly available.

Kevin Huang, Sean Foley, Jihwan Lee et al.

This paper explores the relationship between accent strength and articulatory features inferred from acoustic speech. To quantify accent strength, we compare phonetic transcriptions with transcriptions based on dictionary-based references, computing phoneme-level difference as a measure of accent strength. The proposed framework leverages recent self-supervised learning articulatory inversion techniques to estimate articulatory features. Analyzing a corpus of read speech from American and British English speakers, this study examines correlations between derived articulatory parameters and accent strength proxies, associating systematic articulatory differences with indexed accent strength. Results indicate that tongue positioning patterns distinguish the two dialects, with notable differences inter-dialects in rhotic and low back vowels. These findings contribute to automated accent analysis and articulatory modeling for speech processing applications.

Thanathai Lertpetchpun, Yoonjeong Lee, Jihwan Lee et al.

Accent plays a crucial role in speaker identity and inclusivity in speech technologies. Existing accented text-to-speech (TTS) systems either require large-scale accented datasets or lack fine-grained phoneme-level controllability. We propose a accented TTS framework that combines phonological rules with a multilingual TTS model. The rules are applied to phoneme sequences to transform accent at the phoneme level while preserving intelligibility. The method requires no accented training data and enables explicit phoneme-level accent manipulation. We design rule sets for Spanish- and Indian-accented English, modeling systematic differences in consonants, vowels, and syllable structure arising from phonotactic constraints. We analyze the trade-off between phoneme-level duration alignment and accent as realized in speech timing. Experimental results demonstrate effective accent shift while maintaining speech quality.

Thanathai Lertpetchpun, Thanapat Trachu, Jihwan Lee et al.

Accent is an integral part of society, reflecting multiculturalism and shaping how individuals express identity. The majority of English speakers are non-native (L2) speakers, yet current Text-To-Speech (TTS) systems primarily model American-accented English due limited accented data. We propose \textit{Accent Vector}, a controllable representation that enables accent manipulation in multilingual TTS without requiring accented training data. \textit{Accent Vector} is derived by fine-tuning a TTS system on native speech of a different language (i.e. non-English) and computing task vectors capturing accent characteristics (i.e. in English). By scaling and interpolating the vector, we achieve fine-grained control over accent strength and generate mixed-accent speech. In addition, it generalizes beyond English, enabling accent control across multiple languages. Objective and human evaluations confirm the effectiveness of Accent Vector for fine-grained and compositional accent control.

Jihwan Lee, Parsa Razmara, Kevin Huang et al.

Speech production is a complex process spanning neural planning, motor control, muscle activation, and articulatory kinematics. While the acoustic speech signal is the most accessible product of the speech production act, it does not directly reveal its causal neurophysiological substrates. We present the first simultaneous acquisition of real-time (dynamic) MRI, EEG, and surface EMG, capturing several key aspects of the speech production chain: brain signals, muscle activations, and articulatory movements. This multimodal acquisition paradigm presents substantial technical challenges, including MRI-induced electromagnetic interference and myogenic artifacts. To mitigate these, we introduce an artifact suppression pipeline tailored to this tri-modal setting. Once fully developed, this framework is poised to offer an unprecedented window into speech neuroscience and insights leading to brain-computer interface advances.

Kleanthis Avramidis, Tiantian Feng, Woojae Jeong et al.

Neurophysiological recordings such as electroencephalography (EEG) offer accessible and minimally invasive means of estimating physiological activity for applications in healthcare, diagnostic screening, and even immersive entertainment. However, these recordings yield high-dimensional, noisy time-series data that typically require extensive pre-processing and handcrafted feature extraction to reveal meaningful information. Recently, there has been a surge of interest in applying representation learning techniques from large pre-trained (foundation) models to effectively decode and interpret biosignals. We discuss the challenges posed for incorporating such methods and introduce BioCodec, an alternative representation learning framework inspired by neural codecs to capture low-level signal characteristics in the form of discrete tokens. Pre-trained on thousands of EEG hours, BioCodec shows efficacy across multiple downstream tasks, ranging from clinical diagnostic tasks and sleep physiology to decoding speech and motor imagery, particularly in low-resource settings. Additionally, we provide a qualitative analysis of codebook usage and estimate the spatial coherence of codebook embeddings from EEG connectivity. Notably, we also document the suitability of our method to other biosignal data, i.e., electromyographic (EMG) signals. Overall, the proposed approach provides a versatile solution for biosignal tokenization that performs competitively with state-of-the-art models. The source code and model checkpoints are shared.

Jihwan Lee, Sean Foley, Thanathai Lertpetchpun et al.

We propose ARTI-6, a compact six-dimensional articulatory speech encoding framework derived from real-time MRI data that captures crucial vocal tract regions including the velum, tongue root, and larynx. ARTI-6 consists of three components: (1) a six-dimensional articulatory feature set representing key regions of the vocal tract; (2) an articulatory inversion model, which predicts articulatory features from speech acoustics leveraging speech foundation models, achieving a prediction correlation of 0.87; and (3) an articulatory synthesis model, which reconstructs intelligible speech directly from articulatory features, showing that even a low-dimensional representation can generate natural-sounding speech. Together, ARTI-6 provides an interpretable, computationally efficient, and physiologically grounded framework for advancing articulatory inversion, synthesis, and broader speech technology applications. The source code and speech samples are publicly available.

Dasol Choi, Jihwan Lee, Minjae Lee et al.

While prior research on text-to-image generation has predominantly focused on biases in human depictions, we investigate a more subtle yet pervasive phenomenon: demographic bias in generated objects (e.g., cars). We introduce SODA (Stereotyped Object Diagnostic Audit), a novel framework for systematically measuring such biases. Our approach compares visual attributes of objects generated with demographic cues (e.g., "for young people'') to those from neutral prompts, across 2,700 images produced by three state-of-the-art models (GPT Image-1, Imagen 4, and Stable Diffusion) in five object categories. Through a comprehensive analysis, we uncover strong associations between specific demographic groups and visual attributes, such as recurring color patterns prompted by gender or ethnicity cues. These patterns reflect and reinforce not only well-known stereotypes but also more subtle and unintuitive biases. We also observe that some models generate less diverse outputs, which in turn amplifies the visual disparities compared to neutral prompts. Our proposed auditing framework offers a practical approach for testing, revealing how stereotypes still remain embedded in today's generative models. We see this as an essential step toward more systematic and responsible AI development.

Geonwoo Cho, Jaegyun Im, Jihwan Lee et al.

Generalizing deep reinforcement learning agents to unseen environments remains a significant challenge. One promising solution is Unsupervised Environment Design (UED), a co-evolutionary framework in which a teacher adaptively generates tasks with high learning potential, while a student learns a robust policy from this evolving curriculum. Existing UED methods typically measure learning potential via regret, the gap between optimal and current performance, approximated solely by value-function loss. Building on these approaches, we introduce the transition-prediction error as an additional term in our regret approximation. To capture how training on one task affects performance on others, we further propose a lightweight metric called Co-Learnability. By combining these two measures, we present Transition-aware Regret Approximation with Co-learnability for Environment Design (TRACED). Empirical evaluations show that TRACED produces curricula that improve zero-shot generalization over strong baselines across multiple benchmarks. Ablation studies confirm that the transition-prediction error drives rapid complexity ramp-up and that Co-Learnability delivers additional gains when paired with the transition-prediction error. These results demonstrate how refined regret approximation and explicit modeling of task relationships can be leveraged for sample-efficient curriculum design in UED. Project Page: https://geonwoo.me/traced/

Jihwan Lee, Ruhi Sarikaya, Young-Bum Kim

In this paper, we introduce an approach for leveraging available data across multiple locales sharing the same language to 1) improve domain classification model accuracy in Spoken Language Understanding and user experience even if new locales do not have sufficient data and 2) reduce the cost of scaling the domain classifier to a large number of locales. We propose a locale-agnostic universal domain classification model based on selective multi-task learning that learns a joint representation of an utterance over locales with different sets of domains and allows locales to share knowledge selectively depending on the domains. The experimental results demonstrate the effectiveness of our approach on domain classification task in the scenario of multiple locales with imbalanced data and disparate domain sets. The proposed approach outperforms other baselines models especially when classifying locale-specific domains and also low-resourced domains.

Han Li, Jihwan Lee, Sidharth Mudgal et al.

Domain classification is the task of mapping spoken language utterances to one of the natural language understanding domains in intelligent personal digital assistants (IPDAs). This is a major component in mainstream IPDAs in industry. Apart from official domains, thousands of third-party domains are also created by external developers to enhance the capability of IPDAs. As more domains are developed rapidly, the question of how to continuously accommodate the new domains still remains challenging. Moreover, existing continual learning approaches do not address the problem of incorporating personalized information dynamically for better domain classification. In this paper, we propose CoNDA, a neural network based approach for domain classification that supports incremental learning of new classes. Empirical evaluation shows that CoNDA achieves high accuracy and outperforms baselines by a large margin on both incrementally added new domains and existing domains.

JIhwan Lee, Dongchan Kim, Ruhi Sarikaya et al.

Representation learning is an essential problem in a wide range of applications and it is important for performing downstream tasks successfully. In this paper, we propose a new model that learns coupled representations of domains, intents, and slots by taking advantage of their hierarchical dependency in a Spoken Language Understanding system. Our proposed model learns the vector representation of intents based on the slots tied to these intents by aggregating the representations of the slots. Similarly, the vector representation of a domain is learned by aggregating the representations of the intents tied to a specific domain. To the best of our knowledge, it is the first approach to jointly learning the representations of domains, intents, and slots using their hierarchical relationships. The experimental results demonstrate the effectiveness of the representations learned by our model, as evidenced by improved performance on the contextual cross-domain reranking task.