arXiv subjects

Hsin-Min Wang

Publications and source records attributed to Hsin-Min Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Speech Intelligibility Assessment with Uncertainty-Aware Whisper Embeddings and sLSTM

Non-intrusive speech intelligibility prediction remains challenging due to variability in speakers, noise conditions, and subjective perception. We propose an uncertainty-aware approach that leverages Whisper embeddings in combination with statistical features, specifically the mean, standard deviation, and entropy computed across the embedding dimensions. The entropy, computed via a softmax over the feature dimension, serves as a proxy for uncertainty, complementing global information captured by the mean and standard deviation. To model the sequential structure of speech, we adopt a scalar long short-term memory (sLSTM) network, which efficiently captures long-range dependencies. Building on this foundation, we propose iMTI-Net, an improved multi-target intelligibility prediction network that integrates convolutional neural network (CNN) and sLSTM components within a multitask learning framework. It jointly predicts human intelligibility scores and machine-based word error rates (WER) from Google ASR and Whisper. Experimental results show that iMTI-Net outperforms the original MTI-Net across multiple evaluation metrics, demonstrating the effectiveness of incorporating uncertainty-aware features and the CNN-sLSTM architecture.

eess.AS↗

arXiv2025

A Study on Zero-Shot Non-Intrusive Speech Intelligibility for Hearing Aids Using Large Language Models

This work focuses on zero-shot non-intrusive speech assessment for hearing aids (HA) using large language models (LLMs). Specifically, we introduce GPT-Whisper-HA, an extension of GPT-Whisper, a zero-shot non-intrusive speech assessment model based on LLMs. GPT-Whisper-HA is designed for speech assessment for HA, incorporating MSBG hearing loss and NAL-R simulations to process audio input based on each individual's audiogram, two automatic speech recognition (ASR) modules for audio-to-text representation, and GPT-4o to predict two corresponding scores, followed by score averaging for the final estimated score. Experimental results indicate that GPT-Whisper-HA achieves a 2.59% relative root mean square error (RMSE) improvement over GPT-Whisper, confirming the potential of LLMs for zero-shot speech assessment in predicting subjective intelligibility for HA users.

eess.AS↗

arXiv2025

Improving Perceptual Audio Aesthetic Assessment via Triplet Loss and Self-Supervised Embeddings

We present a system for automatic multi-axis perceptual quality prediction of generative audio, developed for Track 2 of the AudioMOS Challenge 2025. The task is to predict four Audio Aesthetic Scores--Production Quality, Production Complexity, Content Enjoyment, and Content Usefulness--for audio generated by text-to-speech (TTS), text-to-audio (TTA), and text-to-music (TTM) systems. A main challenge is the domain shift between natural training data and synthetic evaluation data. To address this, we combine BEATs, a pretrained transformer-based audio representation model, with a multi-branch long short-term memory (LSTM) predictor and use a triplet loss with buffer-based sampling to structure the embedding space by perceptual similarity. Our results show that this improves embedding discriminability and generalization, enabling domain-robust audio quality assessment without synthetic training data.

eess.AS↗

arXiv2025

NeuroAMP: A Novel End-to-end General Purpose Deep Neural Amplifier for Personalized Hearing Aids

The prevalence of hearing aids is increasing. However, optimizing the amplification processes of hearing aids remains challenging due to the complexity of integrating multiple modular components in traditional methods. To address this challenge, we present NeuroAMP, a novel deep neural network designed for end-to-end, personalized amplification in hearing aids. NeuroAMP leverages both spectral features and the listener's audiogram as inputs, and we investigate four architectures: Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Convolutional Recurrent Neural Network (CRNN), and Transformer. We also introduce Denoising NeuroAMP, an extension that integrates noise reduction along with amplification capabilities for improved performance in real-world scenarios. To enhance generalization, a comprehensive data augmentation strategy was employed during training on diverse speech (TIMIT and TMHINT) and music (Cadenza Challenge MUSIC) datasets. Evaluation using the Hearing Aid Speech Perception Index (HASPI), Hearing Aid Speech Quality Index (HASQI), and Hearing Aid Audio Quality Index (HAAQI) demonstrates that the Transformer architecture within NeuroAMP achieves the best performance, with SRCC scores of 0.9927 (HASQI) and 0.9905 (HASPI) on TIMIT, and 0.9738 (HAAQI) on the Cadenza Challenge MUSIC dataset. Notably, our data augmentation strategy maintains high performance on unseen datasets (e.g., VCTK, MUSDB18-HQ). Furthermore, Denoising NeuroAMP outperforms both the conventional NAL-R+WDRC approach and a two-stage baseline on the VoiceBank+DEMAND dataset, achieving a 10% improvement in both HASPI (0.90) and HASQI (0.59) scores. These results highlight the potential of NeuroAMP and Denoising NeuroAMP to deliver notable improvements in personalized hearing aid amplification.

eess.AS↗

arXiv2025

DRASP: A Dual-Resolution Attentive Statistics Pooling Framework for Automatic MOS Prediction

A pooling mechanism is essential for mean opinion score (MOS) prediction, facilitating the transformation of variable-length audio features into a concise fixed-size representation that effectively encodes speech quality. Existing pooling methods typically operate at a singular granularity, concentrating either on a comprehensive global perspective or a detailed frame-level analysis, which may overlook complementary perceptual insights. To address this limitation, we introduce the Dual-Resolution Attentive Statistics Pooling (DRASP) framework. DRASP integrates both coarse-grained, global statistical summaries and fine-grained, attentive analyses of perceptually significant segments. This dual-view architecture empowers our model to formulate a more thorough and robust representation, capturing both the overarching structural context and salient local details concurrently. Extensive experiments validate the effectiveness and strong generalization ability of the proposed framework. It consistently outperforms various baseline methods across diverse datasets (MusicEval and AES-Natural), MOS prediction backbones (including a CLAP-based model and AudioBox-Aesthetics), and different audio generation systems, achieving a relative improvement of 10.39% in system-level Spearman's rank correlation coefficient (SRCC) over the widely-used average pooling approach.

cs.SD↗

arXiv2025

Revealing the Role of Audio Channels in ASR Performance Degradation

Pre-trained automatic speech recognition (ASR) models have demonstrated strong performance on a variety of tasks. However, their performance can degrade substantially when the input audio comes from different recording channels. While previous studies have demonstrated this phenomenon, it is often attributed to the mismatch between training and testing corpora. This study argues that variations in speech characteristics caused by different recording channels can fundamentally harm ASR performance. To address this limitation, we propose a normalization technique designed to mitigate the impact of channel variation by aligning internal feature representations in the ASR model with those derived from a clean reference channel. This approach significantly improves ASR performance on previously unseen channels and languages, highlighting its ability to generalize across channel and language differences.

cs.SD↗

arXiv2025

QAMRO: Quality-aware Adaptive Margin Ranking Optimization for Human-aligned Assessment of Audio Generation Systems

Evaluating audio generation systems, including text-to-music (TTM), text-to-speech (TTS), and text-to-audio (TTA), remains challenging due to the subjective and multi-dimensional nature of human perception. Existing methods treat mean opinion score (MOS) prediction as a regression problem, but standard regression losses overlook the relativity of perceptual judgments. To address this limitation, we introduce QAMRO, a novel Quality-aware Adaptive Margin Ranking Optimization framework that seamlessly integrates regression objectives from different perspectives, aiming to highlight perceptual differences and prioritize accurate ratings. Our framework leverages pre-trained audio-text models such as CLAP and Audiobox-Aesthetics, and is trained exclusively on the official AudioMOS Challenge 2025 dataset. It demonstrates superior alignment with human evaluations across all dimensions, significantly outperforming robust baseline models.

cs.SD↗

arXiv2025

Feature Importance across Domains for Improving Non-Intrusive Speech Intelligibility Prediction in Hearing Aids

Given the critical role of non-intrusive speech intelligibility assessment in hearing aids (HA), this paper enhances its performance by introducing Feature Importance across Domains (FiDo). We estimate feature importance on spectral and time-domain acoustic features as well as latent representations of Whisper. Importance weights are calculated per frame, and based on these weights, features are projected into new spaces, allowing the model to focus on important areas early. Next, feature concatenation is performed to combine the features before the assessment module processes them. Experimental results show that when FiDo is incorporated into the improved multi-branched speech intelligibility model MBI-Net+, RMSE can be reduced by 7.62% (from 26.10 to 24.11). MBI-Net+ with FiDo also achieves a relative RMSE reduction of 3.98% compared to the best system in the 2023 Clarity Prediction Challenge. These results validate FiDo's effectiveness in enhancing neural speech assessment in HA.

eess.AS↗

arXiv2025

Neuro-MSBG: An End-to-End Neural Model for Hearing Loss Simulation

Hearing loss simulation models are essential for hearing aid deployment. However, existing models have high computational complexity and latency, which limits real-time applications and lack direct integration with speech processing systems. To address these issues, we propose Neuro-MSBG, a lightweight end-to-end model with a personalized audiogram encoder for effective time-frequency modeling. Experiments show that Neuro-MSBG supports parallel inference and retains the intelligibility and perceptual quality of the original MSBG, with a Spearman's rank correlation coefficient (SRCC) of 0.9247 for Short-Time Objective Intelligibility (STOI) and 0.8671 for Perceptual Evaluation of Speech Quality (PESQ). Neuro-MSBG reduces simulation runtime by a factor of 46 (from 0.970 seconds to 0.021 seconds for a 1 second input), further demonstrating its efficiency and practicality.

cs.SD↗

arXiv2025

AVTENet: A Human-Cognition-Inspired Audio-Visual Transformer-Based Ensemble Network for Video Deepfake Detection

The recent proliferation of hyper-realistic deepfake videos has drawn attention to the threat of audio and visual forgeries. Most previous studies on detecting artificial intelligence-generated fake videos only utilize visual modality or audio modality. While some methods exploit audio and visual modalities to detect forged videos, they have not been comprehensively evaluated on multimodal datasets of deepfake videos involving acoustic and visual manipulations, and are mostly based on convolutional neural networks with low detection accuracy. Considering that human cognition instinctively integrates multisensory information including audio and visual cues to perceive and interpret content and the success of transformer in various fields, this study introduces the audio-visual transformer-based ensemble network (AVTENet). This innovative framework tackles the complexities of deepfake technology by integrating both acoustic and visual manipulations to enhance the accuracy of video forgery detection. Specifically, the proposed model integrates several purely transformer-based variants that capture video, audio, and audio-visual salient cues to reach a consensus in prediction. For evaluation, we use the recently released benchmark multimodal audio-video FakeAVCeleb dataset. For a detailed analysis, we evaluate AVTENet, its variants, and several existing methods on multiple test sets of the FakeAVCeleb dataset. Experimental results show that the proposed model outperforms all existing methods and achieves state-of-the-art performance on Testset-I and Testset-II of the FakeAVCeleb dataset. We also compare AVTENet against humans in detecting video forgery. The results show that AVTENet significantly outperforms humans.

cs.CV↗

arXiv2025

HighRateMOS: Sampling-Rate Aware Modeling for Speech Quality Assessment

Modern speech quality prediction models are trained on audio data resampled to a specific sampling rate. When faced with higher-rate audio at test time, these models can produce biased scores. We introduce HighRateMOS, the first non-intrusive mean opinion score (MOS) model that explicitly considers sampling rate. HighRateMOS ensembles three model variants that exploit the following information: (i) a learnable embedding of speech sampling rate, (ii) Wav2vec 2.0 self-supervised embeddings, (iii) multi-scale CNN spectral features, and (iv) MFCC features. In AudioMOS 2025 Track3, HighRateMOS ranked first in five out of eight metrics. Our experiments confirm that modeling the sampling rate directly leads to more robust and sampling-rate-agnostic speech quality predictions.

eess.AS↗

arXiv2025

A Comparative Study on Proactive and Passive Detection of Deepfake Speech

Solutions for defending against deepfake speech fall into two categories: proactive watermarking models and passive conventional deepfake detectors. While both address common threats, their differences in training, optimization, and evaluation prevent a unified protocol for joint evaluation and selecting the best solutions for different cases. This work proposes a framework to evaluate both model types in deepfake speech detection. To ensure fair comparison and minimize discrepancies, all models were trained and tested on common datasets, with performance evaluated using a shared metric. We also analyze their robustness against various adversarial attacks, showing that different models exhibit distinct vulnerabilities to different speech attribute distortions. Our training and evaluation code is available at Github.

cs.SD↗

arXiv2025

A Study on Speech Assessment with Visual Cues

Non-intrusive assessment of speech quality and intelligibility is essential when clean reference signals are unavailable. In this work, we propose a multimodal framework that integrates audio features and visual cues to predict PESQ and STOI scores. It employs a dual-branch architecture, where spectral features are extracted using STFT, and visual embeddings are obtained via a visual encoder. These features are then fused and processed by a CNN-BLSTM with attention, followed by multi-task learning to simultaneously predict PESQ and STOI. Evaluations on the LRS3-TED dataset, augmented with noise from the DEMAND corpus, show that our model outperforms the audio-only baseline. Under seen noise conditions, it improves LCC by 9.61% (0.8397->0.9205) for PESQ and 11.47% (0.7403->0.8253) for STOI. These results highlight the effectiveness of incorporating visual cues in enhancing the accuracy of non-intrusive speech assessment.

eess.AS↗

arXiv2025

A Study on Zero-shot Non-intrusive Speech Assessment using Large Language Models

This work investigates two strategies for zero-shot non-intrusive speech assessment leveraging large language models. First, we explore the audio analysis capabilities of GPT-4o. Second, we propose GPT-Whisper, which uses Whisper as an audio-to-text module and evaluates the naturalness of text via targeted prompt engineering. We evaluate the assessment metrics predicted by GPT-4o and GPT-Whisper, examining their correlation with human-based quality and intelligibility assessments and the character error rate (CER) of automatic speech recognition. Experimental results show that GPT-4o alone is less effective for audio analysis, while GPT-Whisper achieves higher prediction accuracy, has moderate correlation with speech quality and intelligibility, and has higher correlation with CER. Compared to SpeechLMScore and DNSMOS, GPT-Whisper excels in intelligibility metrics, but performs slightly worse than SpeechLMScore in quality estimation. Furthermore, GPT-Whisper outperforms supervised non-intrusive models MOS-SSL and MTI-Net in Spearman's rank correlation for CER of Whisper. These findings validate GPT-Whisper's potential for zero-shot speech assessment without requiring additional training data.

eess.AS↗

arXiv2025

Leveraging Joint Spectral and Spatial Learning with MAMBA for Multichannel Speech Enhancement

In multichannel speech enhancement, effectively capturing spatial and spectral information across different microphones is crucial for noise reduction. Traditional methods, such as CNN or LSTM, attempt to model the temporal dynamics of full-band and sub-band spectral and spatial features. However, these approaches face limitations in fully modeling complex temporal dependencies, especially in dynamic acoustic environments. To overcome these challenges, we modify the current advanced model McNet by introducing an improved version of Mamba, a state-space model, and further propose MCMamba. MCMamba has been completely reengineered to integrate full-band and narrow-band spatial information with sub-band and full-band spectral features, providing a more comprehensive approach to modeling spatial and spectral information. Our experimental results demonstrate that MCMamba significantly improves the modeling of spatial and spectral features in multichannel speech enhancement, outperforming McNet and achieving state-of-the-art performance on the CHiME-3 dataset. Additionally, we find that Mamba performs exceptionally well in modeling spectral information.

eess.AS↗

arXiv2025

HAAQI-Net: A Non-intrusive Neural Music Audio Quality Assessment Model for Hearing Aids

This paper introduces HAAQI-Net, a non-intrusive deep learning-based music audio quality assessment model for hearing aid users. Unlike traditional methods like the Hearing Aid Audio Quality Index (HAAQI) that require intrusive reference signal comparisons, HAAQI-Net offers a more accessible and computationally efficient alternative. By utilizing a Bidirectional Long Short-Term Memory (BLSTM) architecture with attention mechanisms and features extracted from the pre-trained BEATs model, it can predict HAAQI scores directly from music audio clips and hearing loss patterns. Experimental results demonstrate HAAQI-Net's effectiveness, achieving a Linear Correlation Coefficient (LCC) of 0.9368 , a Spearman's Rank Correlation Coefficient (SRCC) of 0.9486 , and a Mean Squared Error (MSE) of 0.0064 and inference time significantly reduces from 62.52 to 2.54 seconds. To address computational overhead, a knowledge distillation strategy was applied, reducing parameters by 75.85% and inference time by 96.46%, while maintaining strong performance (LCC: 0.9071 , SRCC: 0.9307 , MSE: 0.0091 ). To expand its capabilities, HAAQI-Net was adapted to predict subjective human scores like the Mean Opinion Score (MOS) through fine-tuning. This adaptation significantly improved prediction accuracy, validated through statistical analysis. Furthermore, the robustness of HAAQI-Net was evaluated under varying Sound Pressure Level (SPL) conditions, revealing optimal performance at a reference SPL of 65 dB, with accuracy gradually decreasing as SPL deviated from this point. The advancements in subjective score prediction, SPL robustness, and computational efficiency position HAAQI-Net as a scalable solution for music audio quality assessment in hearing aid applications, contributing to efficient and accurate models in audio signal processing and hearing aid technology.

eess.AS↗

arXiv2025

Channel-Aware Domain-Adaptive Generative Adversarial Network for Robust Speech Recognition

While pre-trained automatic speech recognition (ASR) systems demonstrate impressive performance on matched domains, their performance often degrades when confronted with channel mismatch stemming from unseen recording environments and conditions. To mitigate this issue, we propose a novel channel-aware data simulation method for robust ASR training. Our method harnesses the synergistic power of channel-extractive techniques and generative adversarial networks (GANs). We first train a channel encoder capable of extracting embeddings from arbitrary audio. On top of this, channel embeddings are extracted using a minimal amount of target-domain data and used to guide a GAN-based speech synthesizer. This synthesizer generates speech that faithfully preserves the phonetic content of the input while mimicking the channel characteristics of the target domain. We evaluate our method on the challenging Hakka Across Taiwan (HAT) and Taiwanese Across Taiwan (TAT) corpora, achieving relative character error rate (CER) reductions of 20.02% and 9.64%, respectively, compared to the baselines. These results highlight the efficacy of our channel-aware data simulation method for bridging the gap between source- and target-domain acoustics.

cs.SD↗

arXiv2024

Deep Learning-based Non-Intrusive Multi-Objective Speech Assessment Model with Cross-Domain Features

In this study, we propose a cross-domain multi-objective speech assessment model called MOSA-Net, which can estimate multiple speech assessment metrics simultaneously. Experimental results show that MOSA-Net can improve the linear correlation coefficient (LCC) by 0.026 (0.990 vs 0.964 in seen noise environments) and 0.012 (0.969 vs 0.957 in unseen noise environments) in perceptual evaluation of speech quality (PESQ) prediction, compared to Quality-Net, an existing single-task model for PESQ prediction, and improve LCC by 0.021 (0.985 vs 0.964 in seen noise environments) and 0.047 (0.836 vs 0.789 in unseen noise environments) in short-time objective intelligibility (STOI) prediction, compared to STOI-Net (based on CRNN), an existing single-task model for STOI prediction. Moreover, MOSA-Net, originally trained to assess objective scores, can be used as a pre-trained model to be effectively adapted to an assessment model for predicting subjective quality and intelligibility scores with a limited amount of training data. Experimental results show that MOSA-Net can improve LCC by 0.018 (0.805 vs 0.787) in mean opinion score (MOS) prediction, compared to MOS-SSL, a strong single-task model for MOS prediction. In light of the confirmed prediction capability, we further adopt the latent representations of MOSA-Net to guide the speech enhancement (SE) process and derive a quality-intelligibility (QI)-aware SE (QIA-SE) approach accordingly. Experimental results show that QIA-SE provides superior enhancement performance compared with the baseline SE system in terms of objective evaluation metrics and qualitative evaluation test. For example, QIA-SE can improve PESQ by 0.301 (2.953 vs 2.652 in seen noise environments) and 0.18 (2.658 vs 2.478 in unseen noise environments) over a CNN-based baseline SE model.

eess.AS↗