arXiv subjects

Tsun-An Hsieh

Publications and source records attributed to Tsun-An Hsieh.

14 recordsLinked to original sources

arXiv2026

AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking

Internet audio-visual clips convey meaning through time-varying sound and motion, which extend beyond what text alone can represent. To examine whether AI models can understand such signals in human cultural contexts, we introduce AVMeme Exam, a human-curated benchmark of over one thousand iconic Internet sounds and videos spanning speech, songs, music, and sound effects. Each meme is paired with a unique Q&A assessing levels of understanding from surface content to context and emotion to usage and world knowledge, along with metadata such as original year, transcript, summary, and sensitivity. We systematically evaluate state-of-the-art multimodal large language models (MLLMs) alongside human participants using this benchmark. Our results reveal a consistent limitation: current models perform poorly on textless music and sound effects, and struggle to think in context and in culture compared to surface content. These findings highlight a key gap in human-aligned multimodal intelligence and call for models that can perceive contextually and culturally beyond the surface of what they hear and see. Project page: avmemeexam.github.io/public

cs.SD↗

arXiv2025

That's Deprecated! Understanding, Detecting, and Steering Knowledge Conflicts in Language Models for Code Generation

This paper investigates how large language models (LLMs) behave when faced with discrepancies between their parametric knowledge and conflicting information contained in a prompt. Building on prior question-answering (QA) research, we extend the investigation of knowledge conflicts to the realm of code generation. We propose a domain-agnostic framework for constructing and interpreting such conflicts, along with a novel evaluation method and dataset tailored to code conflict scenarios. Our experiments indicate that sufficiently large LLMs encode the notion of a knowledge conflict in their parameters, enabling us to detect knowledge conflicts with up to \textbf{80.65\%} accuracy. Building on these insights, we show that activation-level steering can achieve up to a \textbf{12.6\%} improvement in steering success over a random baseline. However, effectiveness depends critically on balancing model size, task domain, and steering direction. The experiment code and data will be made publicly available after acceptance.

cs.CL↗

arXiv2025

Adaptive Deterministic Flow Matching for Target Speaker Extraction

Generative target speaker extraction (TSE) methods often produce more natural outputs than predictive models. Recent work based on diffusion or flow matching (FM) typically relies on a small, fixed number of reverse steps with a fixed step size. We introduce Adaptive Discriminative Flow Matching TSE (AD-FlowTSE), which extracts the target speech using an adaptive step size. We formulate TSE within the FM paradigm but, unlike prior FM-based speech enhancement and TSE approaches that transport between the mixture (or a normal prior) and the clean-speech distribution, we define the flow between the background and the source, governed by the mixing ratio (MR) of the source and background that creates the mixture. This design enables MR-aware initialization, where the model starts at an adaptive point along the background-source trajectory rather than applying the same reverse schedule across all noise levels. Experiments show that AD-FlowTSE achieves strong TSE with as few as a single step, and that incorporating auxiliary MR estimation further improves target speech accuracy. Together, these results highlight that aligning the transport path with the mixture composition and adapting the step size to noise conditions yields efficient and accurate TSE.

eess.AS↗

arXiv2025

Towards Real-Time Generative Speech Restoration with Flow-Matching

Generative models have shown robust performance on speech enhancement and restoration tasks, but most prior approaches operate offline with high latency, making them unsuitable for streaming applications. In this work, we investigate the feasibility of a low-latency, real-time generative speech restoration system based on flow-matching (FM). Our method tackles diverse real-world tasks, including denoising, dereverberation, and generative restoration. The proposed causal architecture without time-downsampling achieves introduces an total latency of only 20 ms, suitable for real-time communication. In addition, we explore a broad set of architectural variations and sampling strategies to ensure effective training and efficient inference. Notably, our flow-matching model maintains high enhancement quality with only 5 number of function evaluations (NFEs) during sampling, achieving similar performance as when using ~20 NFEs under the same conditions. Experimental results indicate that causal FM-based models favor few-step reverse sampling, and smaller backbones degrade with longer reverse trajectories. We further show a side-by-side comparison of FM to typical adversarial-loss-based training for the same model architecture.

eess.AS↗

arXiv2025

TGIF: Talker Group-Informed Familiarization of Target Speaker Extraction

State-of-the-art target speaker extraction (TSE) systems are typically designed to generalize to any given mixing environment, necessitating a model with a large enough capacity as a generalist. Personalized speech enhancement could be a specialized solution that adapts to single-user scenarios, but it overlooks the practical need for customization in cases where only a small number of talkers are involved, e.g., TSE for a specific family. We address this gap with the proposed concept, talker group-informed familiarization (TGIF) of TSE, where the TSE system specializes in a particular group of users, which is challenging due to the inherent absence of a clean speech target. To this end, we employ a knowledge distillation approach, where a group-specific student model learns from the pseudo-clean targets generated by a large teacher model. This tailors the student model to effectively extract the target speaker from the particular talker group while maintaining computational efficiency. Experimental results demonstrate that our approach outperforms the baseline generic models by adapting to the unique speech characteristics of a given speaker group. Our newly proposed TGIF concept underscores the potential of developing specialized solutions for diverse and real-world applications, such as on-device TSE on a family-owned device.

eess.AS↗

arXiv2024

Multimodal Representation Loss Between Timed Text and Audio for Regularized Speech Separation

Recent studies highlight the potential of textual modalities in conditioning the speech separation model's inference process. However, regularization-based methods remain underexplored despite their advantages of not requiring auxiliary text data during the test time. To address this gap, we introduce a timed text-based regularization (TTR) method that uses language model-derived semantics to improve speech separation models. Our approach involves two steps. We begin with two pretrained audio and language models, WavLM and BERT, respectively. Then, a Transformer-based audio summarizer is learned to align the audio and word embeddings and to minimize their gap. The summarizer Transformer, incorporated as a regularizer, promotes the separated sources' alignment with the semantics from the timed text. Experimental results show that the proposed TTR method consistently improves the various objective metrics of the separation results over the unregularized baselines.

eess.AS↗

arXiv2024

On the Importance of Neural Wiener Filter for Resource Efficient Multichannel Speech Enhancement

We introduce a time-domain framework for efficient multichannel speech enhancement, emphasizing low latency and computational efficiency. This framework incorporates two compact deep neural networks (DNNs) surrounding a multichannel neural Wiener filter (NWF). The first DNN enhances the speech signal to estimate NWF coefficients, while the second DNN refines the output from the NWF. The NWF, while conceptually similar to the traditional frequency-domain Wiener filter, undergoes a training process optimized for low-latency speech enhancement, involving fine-tuning of both analysis and synthesis transforms. Our research results illustrate that the NWF output, having minimal nonlinear distortions, attains performance levels akin to those of the first DNN, deviating from conventional Wiener filter paradigms. Training all components jointly outperforms sequential training, despite its simplicity. Consequently, this framework achieves superior performance with fewer parameters and reduced computational demands, making it a compelling solution for resource-efficient multichannel speech enhancement.

cs.SD↗

arXiv2022

Inference and Denoise: Causal Inference-based Neural Speech Enhancement

This study addresses the speech enhancement (SE) task within the causal inference paradigm by modeling the noise presence as an intervention. Based on the potential outcome framework, the proposed causal inference-based speech enhancement (CISE) separates clean and noisy frames in an intervened noisy speech using a noise detector and assigns both sets of frames to two mask-based enhancement modules (EMs) to perform noise-conditional SE. Specifically, we use the presence of noise as guidance for EM selection during training, and the noise detector selects the enhancement module according to the prediction of the presence of noise for each frame. Moreover, we derived a SE-specific average treatment effect to quantify the causal effect adequately. Experimental evidence demonstrates that CISE outperforms a non-causal mask-based SE approach in the studied settings and has better performance and efficiency than more complex SE models.

eess.AS↗

arXiv2022

Speech Recovery for Real-World Self-powered Intermittent Devices

The incompleteness of speech inputs severely degrades the performance of all the related speech signal processing applications. Although many researches have been proposed to address this issue, they controlled the data missing conditions by simulation with self-defined masking lengths or sizes. Besides, the masking definitions are different among all these experimental settings. This paper presents a novel intermittent speech recovery (ISR) system for real-world self-powered intermittent devices. Three contributive stages: interpolation, enhancement, and combination are applied to the ISR system for speech reconstruction. The experimental results show that our recovery system increases speech quality by up to 591.7%, while increasing speech intelligibility by up to 80.5%. Most importantly, the proposed ISR system improves the WER scores by up to 52.6%. The promising results not only confirm the effectiveness of the reconstruction but also encourage the utilization of these battery-free wearable/IoT devices.

cs.SD↗

arXiv2021

OSSEM: one-shot speaker adaptive speech enhancement using meta learning

Although deep learning (DL) has achieved notable progress in speech enhancement (SE), further research is still required for a DL-based SE system to adapt effectively and efficiently to particular speakers. In this study, we propose a novel meta-learning-based speaker-adaptive SE approach (called OSSEM) that aims to achieve SE model adaptation in a one-shot manner. OSSEM consists of a modified transformer SE network and a speaker-specific masking (SSM) network. In practice, the SSM network takes an enrolled speaker embedding extracted using ECAPA-TDNN to adjust the input noisy feature through masking. To evaluate OSSEM, we designed a modified Voice Bank-DEMAND dataset, in which one utterance from the testing set was used for model adaptation, and the remaining utterances were used for testing the performance. Moreover, we set restrictions allowing the enhancement process to be conducted in real time, and thus designed OSSEM to be a causal SE system. Experimental results first show that OSSEM can effectively adapt a pretrained SE model to a particular speaker with only one utterance, thus yielding improved SE results. Meanwhile, OSSEM exhibits a competitive performance compared to state-of-the-art causal SE systems.

eess.AS↗

arXiv2021

MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement

The discrepancy between the cost function used for training a speech enhancement model and human auditory perception usually makes the quality of enhanced speech unsatisfactory. Objective evaluation metrics which consider human perception can hence serve as a bridge to reduce the gap. Our previously proposed MetricGAN was designed to optimize objective metrics by connecting the metric with a discriminator. Because only the scores of the target evaluation functions are needed during training, the metrics can even be non-differentiable. In this study, we propose a MetricGAN+ in which three training techniques incorporating domain-knowledge of speech processing are proposed. With these techniques, experimental results on the VoiceBank-DEMAND dataset show that MetricGAN+ can increase PESQ score by 0.3 compared to the previous MetricGAN and achieve state-of-the-art results (PESQ score = 3.15).

cs.SD↗

arXiv2021

Improving Perceptual Quality by Phone-Fortified Perceptual Loss using Wasserstein Distance for Speech Enhancement

Speech enhancement (SE) aims to improve speech quality and intelligibility, which are both related to a smooth transition in speech segments that may carry linguistic information, e.g. phones and syllables. In this study, we propose a novel phone-fortified perceptual loss (PFPL) that takes phonetic information into account for training SE models. To effectively incorporate the phonetic information, the PFPL is computed based on latent representations of the wav2vec model, a powerful self-supervised encoder that renders rich phonetic information. To more accurately measure the distribution distances of the latent representations, the PFPL adopts the Wasserstein distance as the distance measure. Our experimental results first reveal that the PFPL is more correlated with the perceptual evaluation metrics, as compared to signal-level losses. Moreover, the results showed that the PFPL can enable a deep complex U-Net SE model to achieve highly competitive performance in terms of standardized quality and intelligibility evaluations on the Voice Bank-DEMAND dataset.

cs.SD↗

arXiv2021

Boosting Objective Scores of a Speech Enhancement Model by MetricGAN Post-processing

The Transformer architecture has demonstrated a superior ability compared to recurrent neural networks in many different natural language processing applications. Therefore, our study applies a modified Transformer in a speech enhancement task. Specifically, positional encoding in the Transformer may not be necessary for speech enhancement, and hence, it is replaced by convolutional layers. To further improve the perceptual evaluation of the speech quality (PESQ) scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned using a MetricGAN framework. The proposed MetricGAN can be treated as a general post-processing module to further boost the objective scores of interest. The experiments were conducted using the data sets provided by the organizer of the Deep Noise Suppression (DNS) challenge. Experimental results demonstrated that the proposed system outperformed the challenge baseline, in both subjective and objective evaluations, with a large margin.

eess.AS↗

arXiv2020

WaveCRN: An Efficient Convolutional Recurrent Neural Network for End-to-end Speech Enhancement

Due to the simple design pipeline, end-to-end (E2E) neural models for speech enhancement (SE) have attracted great interest. In order to improve the performance of the E2E model, the locality and temporal sequential properties of speech should be efficiently taken into account when modelling. However, in most current E2E models for SE, these properties are either not fully considered or are too complex to be realized. In this paper, we propose an efficient E2E SE model, termed WaveCRN. In WaveCRN, the speech locality feature is captured by a convolutional neural network (CNN), while the temporal sequential property of the locality feature is modeled by stacked simple recurrent units (SRU). Unlike a conventional temporal sequential model that uses a long short-term memory (LSTM) network, which is difficult to parallelize, SRU can be efficiently parallelized in calculation with even fewer model parameters. In addition, in order to more effectively suppress the noise components in the input noisy speech, we derive a novel restricted feature masking (RFM) approach that performs enhancement on the feature maps in the hidden layers; this is different from the approach that applies the estimated ratio mask on the noisy spectral features, which is commonly used in speech separation methods. Experimental results on speech denoising and compressed speech restoration tasks confirm that with the lightweight architecture of SRU and the feature-mapping-based RFM, WaveCRN performs comparably with other state-of-the-art approaches with notably reduced model complexity and inference time.

eess.AS↗