We present the first edition of the VoiceMOS Challenge, a scientific event that aims to promote the study of automatic prediction of the mean opinion score (MOS) of synthetic speech. This challenge drew 22 participating teams from academia and industry who tried a variety of approaches to tackle the problem of predicting human ratings of synthesized speech. The listening test data for the main track of the challenge consisted of samples from 187 different text-to-speech and voice conversion systems spanning over a decade of research, and the out-of-domain track consisted of data from more recent systems rated in a separate listening test. Results of the challenge show the effectiveness of fine-tuning self-supervised speech models for the MOS prediction task, as well as the difficulty of predicting MOS ratings for unseen speakers and listeners, and for unseen systems in the out-of-domain setting.
The VoiceMOS Challenge 2026: Evaluating Speech Enhancement, Emotional TTS and Accented TTS Systems
Nagoya University(名古屋大学);Shanghai Jiao Tong University(上海交通大学);Technische Universität Braunschweig(布伦瑞克工业大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));University of Edinburgh(爱丁堡大学);National Institute of Information and Communications Technology(日本国立信息通信技术研究所)
The VoiceMOS Challenge 2024: Beyond Speech Quality Prediction
Nagoya University(名古屋大学);NVIDIA(英伟达);National Institute Of Information And Communications Technology(日本信息通信研究机构);Academia Sinica(中央研究院);National Institute of Informatics(情报信息学研究所)
Predictions of Subjective Ratings and Spoofing Assessments of Voice Conversion Challenge 2020 Submissions
National University of Singapore(新加坡国立大学);University of Eastern Finland(东芬兰大学);Nagoya University(名古屋大学);University of Science and Technology of China(中国科学技术大学);National Institute of Informatics(国立信息学研究所)
The T05 System for The VoiceMOS Challenge 2024: Transfer Learning from Deep Image Classifier to Naturalness MOS Prediction of High-Quality Synthetic Speech
An Extensive Analysis of the Singing Voice Conversion Challenge 2025 Evaluation Results
Graduate School of Informatics, Nagoya University, Japan(名古屋大学信息学研究科,日本);Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳));Carnegie Mellon University, USA(卡内基梅隆大学,美国);National Institute of Informatics, Japan(日本信息处理学会);Information Technology Center, Nagoya University, Japan(名古屋大学信息技术中心,日本)
Evaluation of the Speech Resynthesis Capabilities of the VoicePrivacy Challenge Baseline B1
International Audio Laboratories Erlangen(埃尔朗根国际音频实验室);Friedrich-Alexander-Universität Erlangen-Nürnberg(弗里德里希-亚历山大-埃尔朗根-纽伦堡大学);University of Erlangen-Nürnberg(埃尔朗根-纽伦堡大学);Fraunhofer IIS(弗劳恩霍夫集成电路研究所)
Voice Conversion Challenge 2020: Intra-lingual semi-parallel and cross-lingual voice conversion
National Institute of Informatics(国立信息学研究所);Nagoya University(名古屋大学);National University of Singapore(新加坡国立大学);University of Eastern Finland(东芬兰大学);University of Science and Technology of China(中国科学技术大学)
A Comparative Study of Voice Conversion Models with Large-Scale Speech and Singing Data: The T13 Systems for the Singing Voice Conversion Challenge 2023
MOSNet: Deep Learning based Objective Assessment for Voice Conversion
Institute of Information Science, Academia Sinica(中央研究院资讯科学研究所);Research Center for Information Technology Innovation, Academia Sinica(中央研究院资讯科技创新研究中心);National Institute of Informatics(国立情报学研究所)
Does a PESQNet (Loss) Require a Clean Reference Input? The Original PESQ Does, But ACR Listening Tests Don't
语音质量的感知评价(PESQ)需要一个干净的语音参考作为输入,但预测的结果(无参考)绝对类别评级(ACR)测试。在这项工作中,我们训练一个完全卷积的递归神经网络(FCRN)作为深度噪声抑制(DNS)模型,使用非侵入式或侵入式PESQNet,其中只有后者可以访问干净的语音参考。PESQNet被用作在DNS训练期间提供感知损失的中介,以最大化增强语音信号的PESQ分数。对于侵入式PESQNet,我们研究了两种拓扑结构,称为早期融合(EF)和中期融合(MF)PESQNet,并与非侵入式PESQNet进行比较,以评估和量化在DNS训练期间采用干净语音参考输入的好处。详细的分析表明,使用MF侵入式PESQNet训练的DNS分别比Interspeech 2021 DNS Challenge基线和使用MSE损失训练的相同DNS分别高出0.23和0.12个PESQ点。此外,我们可以证明,与使用非侵入式PESQNet训练的DNS相比,只获得了边际效益。因此,作为ACR听力测试,PESQNet不一定需要干净的语音参考输入,从而可以使用真实数据进行DNS训练。
Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation
Barcelona Supercomputing Center, Spain(巴塞罗那超级计算中心);Universitat Politècnica de Catalunya, Spain(加泰罗尼亚理工大学);DFKI GmbH, Saarland Informatics Campus, Saarbrücken, Germany(DFKI GmbH萨尔兰信息技术校区)
Sequential Skip Prediction with Few-shot in Streamed Music Contents
Seoul National University(首尔国立大学);Institute for Industrial Systems Innovation(工业系统创新研究所);Center for Super Intelligence(超级智能中心);Music and Audio Research Group(音乐与音频研究组)
本文概述了为WSDM Cup 2019 Spotify顺序跳过预测挑战赛(团队名称:mimbres)提交的算法。在挑战中,提供了完整的信息,包括收听会话的前半部分的声学特征和用户交互日志。我们的目标是预测是否会跳过会话后半部分中的单个曲目,仅给定声学特征。我们提出了两种不同的算法,分别基于度量学习和序列学习。实验结果表明,序列学习方法的性能明显优于度量学习方法。此外,我们进行了额外的实验,发现使用完整的用户日志信息可以实现显着的性能增益。
The NVVSpeech Challenge 2026: Overview and Lessons for Speech Understanding and Generation with Nonverbal Vocalizations
School of Intelligence Science and Technology, Nanjing University(南京大学智能科学与技术学院);School of Computer Science, Northwestern Polytechnical University(西北工业大学计算机学院);Shenzhen Pimei Technology(深圳湃美科技);National Taiwan University(台湾大学);Tongji University(同济大学);Hong Kong University of Science and Technology(香港科技大学);Yutuzhineng(语知能);AISHELL;State Key Laboratory of Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室)
The GENEA Challenge 2026: A Large-Scale Disentangled Evaluation of Speech-Driven Gesture Generation on the Seamless Interaction Dataset
KTH Royal Institute of Technology(瑞典皇家理工学院);Bielefeld University(比勒费尔德大学);National Library of Sweden(瑞典国家图书馆);Electronics and Telecommunications Research Institute (ETRI)(电子通信研究院);Motorica AB(Motorica AB公司)