arXiv subjects

Wen-Chin Huang

Publications and source records attributed to Wen-Chin Huang.

At least 19 recordsLinked to original sources

arXiv2026

The ISCSLP 2026 Real-World Audio-Visual Speech Enhancement Challenge: A Benchmark with Natural Mixtures and Degraded Video

Audio-visual speech enhancement (AVSE) uses a target speaker's visible articulation to recover that speaker's voice from overlapped speech. Yet most evaluation protocols rely on synthetic mixtures and reliable video. The ISCSLP 2026 Real-World AVSE Challenge addresses both gaps. Track~1 combines naturally recorded two-talker mixtures, which lack a clean reference, with reference-available remixes of the same speakers; Track~2 additionally degrades the target video in five ways and adds 3-m far-field recordings. Sixteen and twelve teams were ranked on speaker-disjoint test data by rank averaging over waveform fidelity, predicted quality, transcription accuracy, and speaker similarity. On Track~1 remixes, the best system reaches 12.7~dB SI-SDR and 0.85 STOI, but natural recordings remain harder: even the lowest CER rises from 9.0\% to 14.7\%. The leading systems use video mainly for speaker attribution rather than signal reconstruction, and the top two lose under 0.5~dB SI-SDR on Track~2. UTMOS and DNSMOS rank systems differently from the other metrics, so no single metric captures target-speech recovery. We release the baselines, evaluator, and official results.

eess.AS↗

arXiv2026

NeuMark: Neural Codec Resynthesis-Robust Audio Watermarking in the Codec Latent Space

Audio watermarking is increasingly important for tracing generated speech. Several audio watermarking methods have been proposed to embed the watermark in various domains, such as waveform, timbre feature, or latent representations, for making the embedded watermark robust against traditional digital signal processing (DSP) attacks. On the other hand, modern neural codecs introduce a different threat from DSP attacks: they resynthesize speech through quantized acoustic representations and can remove the embedded watermark evi- dence that is not aligned with codec-preserved structure. In this paper, we propose NeuMark, a codec-latent audio watermarking framework that embeds watermark evidence into SpeechTok- enizer acoustic tokens to address this resynthesis threat. NeuMark uses cross-attention to inject a 16-bit message across residual vector quantization (RVQ) layers, distributing the watermark over codec-aligned latent structure. Experimental results show that NeuMark substantially improves robustness under neural- codec resynthesis while supporting both watermark detection and message recovery. We also analyze the trade-off between reconstruction-referenced transparency and original-referenced robustness.

cs.SD↗

arXiv2026

Attacking UTMOS: Probing the Robustness of a Speech Quality Assessment Model

UTMOS has become one of the most commonly used deep neural network-based speech quality assessment (SQA) metrics in speech processing research. In this paper, we attack UTMOS to probe its robustness. Starting from high-quality speech samples, we optimize the input in two directions: a score-preserving attack, which degrades perceived quality while maintaining the predicted score, and a quality-preserving attack, which lowers the predicted score while maintaining perceived quality. We consider three input spaces: raw waveform, mel spectrogram with a HiFi-GAN vocoder, and the latent space of EnCodec, a neural audio codec. Experimental results show that score-preserving attacks are effective against UTMOS. Although perfect quality-preserving attacks are more difficult, optimization in the EnCodec latent space provides the best chance of success. These results reveal failure modes of UTMOS and highlight the importance of robustness analysis for DNN-based SQA metrics.

cs.SD↗

arXiv2026

The VoiceMOS Challenge 2026: Evaluating Speech Enhancement, Emotional TTS and Accented TTS Systems

We present the results of the VoiceMOS Challenge 2026, the fifth edition of a scientific challenge on automatic prediction of subjective speech assessments. After expanding the scope to music and general audio in 2025, we refocused the evaluation target on speech and organized three tracks: prediction of absolute and comparative category ratings for enhanced speech, prediction of naturalness and emotion-related tasks for emotional text-to-speech systems, and prediction of speaker and accent similarity for codec-based speech synthesis systems. The challenge attracted a total of 18 teams worldwide, with most teams successfully surpassing the provided baselines. We summarize the challenge results, representative top-performing systems, participant feedback, and directions for future editions.

cs.SD↗

arXiv2026

An Extensive Analysis of the Singing Voice Conversion Challenge 2025 Evaluation Results

We present a thorough analysis of the findings of the latest iteration of the Singing Voice Conversion Challenge, a scientific event aiming to compare and understand different voice conversion systems in a controlled environment. Compared to previous iterations which solely focused on converting the singer identity, this year we also focused on converting the singing style of the singer. To create a controlled environment and thorough evaluations, we developed a new challenge database, introduced two tasks, open-sourced baselines, and conducted large-scale crowd-sourced listening tests and objective evaluations. The challenge was run for two months and in total we evaluated 33 different systems. The results of the large-scale crowd-sourced listening test showed that top systems had comparable singer identity scores to ground truth samples. However, modeling the singing style and consequently achieving high naturalness still remains a challenge in this task, primarily due to the difficulty in modeling dynamic information in breathy, glissando, and vibrato singing styles. Further analyses of the challenge also discuss the limitations of both the traditional similarity test and the dynamic preference test in evaluating singing style similarity. Moreover, calculating Spearman's rank correlation coefficient shows that dependent objective metrics such as chroma-alignment and non-match metrics such as speaker embeddings are the most correlated to subjective scores, but are still not at a level where it could be considered as a true replacement for subjective scores.

cs.SD↗

arXiv2026

MOS-Bench: Benchmarking Generalization Abilities of Subjective Speech Quality Assessment Models

In this paper, we study the task of subjective speech quality assessment (SSQA), which refers to predicting the perceptual quality of speech. Owing to the development of deep neural network models, SSQA has greatly advanced and has been widely applied in scientific papers to evaluate speech generation systems. Nonetheless, the insufficient out-of-domain (OOD) generalization ability of current SSQA models is underexplored and often overlooked by researchers. To study this problem systematically, we present MOS-Bench, a diverse SSQA dataset collection that currently contains 8 training sets and 17 test sets. Through extensive experiments, we first highlight the OOD generalization challenges of existing models. We then evaluate the efficacy of multiple-dataset training, comparing straightforward data pooling against AlignNet, an existing domain-aware method. We demonstrate that pooling multiple training sets provides a simple yet effective solution, and variation in the data is a key factor for robust generalization beyond training data size.

cs.SD↗

arXiv2026

MOS-Bias: From Hidden Gender Bias to Gender-Aware Speech Quality Assessment

The Mean Opinion Score (MOS) serves as the standard metric for speech quality assessment, yet biases in human annotations remain underexplored. We conduct the first systematic analysis of gender bias in MOS, revealing that male listeners consistently assign higher scores than female listeners--a gap that is most pronounced in low-quality speech and gradually diminishes as quality improves. This quality-dependent structure proves difficult to eliminate through simple calibration. We further demonstrate that automated MOS models trained on aggregated labels exhibit predictions skewed toward male standards of perception. To address this, we propose a gender-aware model that learns gender-specific scoring patterns through abstracting binary group embeddings, thereby improving overall and gender-specific prediction accuracy. This study establishes that gender bias in MOS constitutes a systematic, learnable pattern demanding attention in equitable speech evaluation.

eess.AS↗

arXiv2026

CodecMOS-Accent: A MOS Benchmark of Resynthesized and TTS Speech from Neural Codecs Across English Accents

We present the CodecMOS-Accent dataset, a mean opinion score (MOS) benchmark designed to evaluate neural audio codec (NAC) models and the large language model (LLM)-based text-to-speech (TTS) models trained upon them, especially across non-standard speech like accented speech. The dataset comprises 4,000 codec resynthesis and TTS samples from 24 systems, featuring 32 speakers spanning ten accents. A large-scale subjective test was conducted to collect 19,600 annotations from 25 listeners across three dimensions: naturalness, speaker similarity, and accent similarity. This dataset does not only represent an up-to-date study of recent speech synthesis system performance but reveals insights including a tight relationship between speaker and accent similarity, the predictive power of objective metrics, and a perceptual bias when listeners share the same accent with the speaker. This dataset is expected to foster research on more human-centric evaluation for NAC and accented TTS.

cs.SD↗

arXiv2025

The AudioMOS Challenge 2025

This is the summary paper for the AudioMOS Challenge 2025, the very first challenge for automatic subjective quality prediction for synthetic audio. The challenge consists of three tracks. The first track aims to assess text-to-music samples in terms of overall quality and textual alignment. The second track is based on the four evaluation dimensions of Meta Audiobox Aesthetics, and the test set consists of text-to-speech, text-to-audio, and text-to-music samples. The third track focuses on synthetic speech quality assessment in different sampling rates. The challenge attracted 24 unique teams from both academia and industry, and improvements over the baselines were confirmed. The outcome of this challenge is expected to facilitate development and progress in the field of automatic evaluation for audio generation systems.

cs.SD↗

arXiv2025

Advancing Speech Quality Assessment Through Scientific Challenges and Open-source Activities

Speech quality assessment (SQA) refers to the evaluation of speech quality, and developing an accurate automatic SQA method that reflects human perception has become increasingly important, in order to keep up with the generative AI boom. In recent years, SQA has progressed to a point that researchers started to faithfully use automatic SQA in research papers as a rigorous measurement of goodness for speech generation systems. We believe that the scientific challenges and open-source activities of late have stimulated the growth in this field. In this paper, we review recent challenges as well as open-source implementations and toolkits for SQA, and highlight the importance of maintaining such activities to facilitate the development of not only SQA itself but also generative AI for speech.

cs.SD↗

arXiv2025

Serenade: A Singing Style Conversion Framework Based On Audio Infilling

We propose Serenade, a novel framework for the singing style conversion (SSC) task. Although singer identity conversion has made great strides in the previous years, converting the singing style of a singer has been an unexplored research area. We find three main challenges in SSC: modeling the target style, disentangling source style, and retaining the source melody. To model the target singing style, we use an audio infilling task by predicting a masked segment of the target mel-spectrogram with a flow-matching model using the complement of the masked target mel-spectrogram along with disentangled acoustic features. On the other hand, to disentangle the source singing style, we use a cyclic training approach, where we use synthetic converted samples as source inputs and reconstruct the original source mel-spectrogram as a target. Finally, to retain the source melody better, we investigate a post-processing module using a source-filter-based vocoder and resynthesize the converted waveforms using the original F0 patterns. Our results showed that the Serenade framework can handle generalized SSC tasks with the best overall similarity score, especially in modeling breathy and mixed singing styles. We also found that resynthesizing with the original F0 patterns alleviated out-of-tune singing and improved naturalness, but found a slight tradeoff in similarity due to not changing the F0 patterns into the target style.

cs.SD↗

arXiv2025

Mitigating Language Mismatch in SSL-Based Speaker Anonymization

Speaker anonymization aims to protect speaker identity while preserving content information and the intelligibility of speech. However, most speaker anonymization systems (SASs) are developed and evaluated using only English, resulting in degraded utility for other languages. This paper investigates language mismatch in SASs for Japanese and Mandarin speech. First, we fine-tune a self-supervised learning (SSL)-based content encoder with Japanese speech to verify effective language adaptation. Then, we propose fine-tuning a multilingual SSL model with Japanese speech and evaluating the SAS in Japanese and Mandarin. Downstream experiments show that fine-tuning an English-only SSL model with the target language enhances intelligibility while maintaining privacy and that multilingual SSL further extends SASs' utility across different languages. These findings highlight the importance of language adaptation and multilingual pre-training of SSLs for robust multilingual speaker anonymization.

eess.AS↗

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

SHEET: A Multi-purpose Open-source Speech Human Evaluation Estimation Toolkit

We introduce SHEET, a multi-purpose open-source toolkit designed to accelerate subjective speech quality assessment (SSQA) research. SHEET stands for the Speech Human Evaluation Estimation Toolkit, which focuses on data-driven deep neural network-based models trained to predict human-labeled quality scores of speech samples. SHEET provides comprehensive training and evaluation scripts, multi-dataset and multi-model support, as well as pre-trained models accessible via Torch Hub and HuggingFace Spaces. To demonstrate its capabilities, we re-evaluated SSL-MOS, a speech self-supervised learning (SSL)-based SSQA model widely used in recent scientific papers, on an extensive list of speech SSL models. Experiments were conducted on two representative SSQA datasets named BVCC and NISQA, and we identified the optimal speech SSL model, whose performance surpassed the original SSL-MOS implementation and was comparable to state-of-the-art methods.

cs.SD↗

arXiv2025

Music Similarity Representation Learning Focusing on Individual Instruments with Source Separation and Human Preference

This paper proposes music similarity representation learning (MSRL) based on individual instrument sounds (InMSRL) utilizing music source separation (MSS) and human preference without requiring clean instrument sounds during inference. We propose three methods that effectively improve performance. First, we introduce end-to-end fine-tuning (E2E-FT) for the Cascade approach that sequentially performs MSS and music similarity feature extraction. E2E-FT allows the model to minimize the adverse effects of a separation error on the feature extraction. Second, we propose multi-task learning for the Direct approach that directly extracts disentangled music similarity features using a single music similarity feature extractor. Multi-task learning, which is based on the disentangled music similarity feature extraction and MSS based on reconstruction with disentangled music similarity features, further enhances instrument feature disentanglement. Third, we employ perception-aware fine-tuning (PAFT). PAFT utilizes human preference, allowing the model to perform InMSRL aligned with human perceptual similarity. We conduct experimental evaluations and demonstrate that 1) E2E-FT for Cascade significantly improves InMSRL performance, 2) the multi-task learning for Direct is also helpful to improve disentanglement performance in the feature extraction, 3) PAFT significantly enhances the perceptual InMSRL performance, and 4) Cascade with E2E-FT and PAFT outperforms Direct with the multi-task learning and PAFT.

cs.SD↗

arXiv2024

The VoiceMOS Challenge 2024: Beyond Speech Quality Prediction

We present the third edition of the VoiceMOS Challenge, a scientific initiative designed to advance research into automatic prediction of human speech ratings. There were three tracks. The first track was on predicting the quality of ``zoomed-in'' high-quality samples from speech synthesis systems. The second track was to predict ratings of samples from singing voice synthesis and voice conversion with a large variety of systems, listeners, and languages. The third track was semi-supervised quality prediction for noisy, clean, and enhanced speech, where a very small amount of labeled training data was provided. Among the eight teams from both academia and industry, we found that many were able to outperform the baseline systems. Successful techniques included retrieval-based methods and the use of non-self-supervised representations like spectrograms and pitch histograms. These results showed that the challenge has advanced the field of subjective speech rating prediction.

cs.SD↗

arXiv2024

Quantifying the effect of speech pathology on automatic and human speaker verification

This study investigates how surgical intervention for speech pathology (specifically, as a result of oral cancer surgery) impacts the performance of an automatic speaker verification (ASV) system. Using two recently collected Dutch datasets with parallel pre and post-surgery audio from the same speaker, NKI-OC-VC and SPOKE, we assess the extent to which speech pathology influences ASV performance, and whether objective/subjective measures of speech severity are correlated with the performance. Finally, we carry out a perceptual study to compare judgements of ASV and human listeners. Our findings reveal that pathological speech negatively affects ASV performance, and the severity of the speech is negatively correlated with the performance. There is a moderate agreement in perceptual and objective scores of speaker similarity and severity, however, we could not clearly establish in the perceptual study, whether the same phenomenon also exists in human perception.

cs.SD↗

arXiv2024

A Large-Scale Evaluation of Speech Foundation Models

The foundation model paradigm leverages a shared foundation model to achieve state-of-the-art (SOTA) performance for various tasks, requiring minimal downstream-specific modeling and data annotation. This approach has proven crucial in the field of Natural Language Processing (NLP). However, the speech processing community lacks a similar setup to explore the paradigm systematically. In this work, we establish the Speech processing Universal PERformance Benchmark (SUPERB) to study the effectiveness of the paradigm for speech. We propose a unified multi-tasking framework to address speech processing tasks in SUPERB using a frozen foundation model followed by task-specialized, lightweight prediction heads. Combining our results with community submissions, we verify that the foundation model paradigm is promising for speech, and our multi-tasking framework is simple yet effective, as the best-performing foundation model shows competitive generalizability across most SUPERB tasks. For reproducibility and extensibility, we have developed a long-term maintained platform that enables deterministic benchmarking, allows for result sharing via an online leaderboard, and promotes collaboration through a community-driven benchmark database to support new development cycles. Finally, we conduct a series of analyses to offer an in-depth understanding of SUPERB and speech foundation models, including information flows across tasks inside the models, the correctness of the weighted-sum benchmarking protocol and the statistical significance and robustness of the benchmark.

eess.AS↗