语音/音频处理学术速递[8.21]

今日论文合集：cs.SD语音21篇，eess.AS音频处理21篇。

cs.SD语音

【1】 Audiovisual Moments in Time: A Large-Scale Annotated Dataset of Audiovisual Actions标题：时间中的视听时刻：视听动作的大规模注释数据集链接：https://arxiv.org/abs/2308.09685

作者：Michael Joannou,Pia Rotshtein,Uta Noppeney机构：Computational Neuroscience and Cognitive Robotics Centre, University of Birmingham, Birmingham, UK, University of Haifa, Mount Carmel, Haifa, Israel, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands

【2】 Compensating Removed Frequency Components: Thwarting Voice Spectrum Reduction Attacks标题：补偿去除的频率分量：挫败语音频谱降低攻击链接：https://arxiv.org/abs/2308.09546

作者：Shu Wang,Kun Sun,Qi Li机构：George Mason University, Tsinghua University备注：Accepted by 2024 Network and Distributed System Security Symposium (NDSS'24)

【3】 Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learning标题：Space LibriSpeech：一种用于空间音频学习的扩充数据集链接：https://arxiv.org/abs/2308.09514

作者：Miguel Sarabia,Elena Menyaylenko,Alessandro Toso,Skyler Seto,Zakaria Aldeneh,Shadi Pirhosseinloo,Luca Zappella,Barry-John Theobald,Nicholas Apostoloff,Jonathan Sheaffer机构：Apple备注：None

【4】 Exploring Sampling Techniques for Generating Melodies with a Transformer Language Model标题：探索用Transformer语言模型生成旋律的采样技术链接：https://arxiv.org/abs/2308.09454

作者：Mathias Rose Bjare,Stefan Lattner,Gerhard Widmer机构：Institute of Computational Perception, Johannes Kepler University Linz, Austria, Sony Computer Science Laboratories (CSL), Paris, France, LIT AI Lab, Linz Institute of Technology, Austria备注：7 pages, 5 figures, 1 table, accepted at the 24th Int. Society for Music Information Retrieval Conf., Milan, Italy, 2023

【5】 TrOMR:Transformer-Based Polyphonic Optical Music Recognition标题：TrOMR：基于Transformer的复调光学音乐识别链接：https://arxiv.org/abs/2308.09370

作者：Yixuan Li,Huaping Liu,Qiang Jin,Miaomiao Cai,Peng Li机构：NetEase Cloud Music, Hangzhou, China备注：None

【6】 Lip Reading for Low-resource Languages by Learning and Combining General Speech Knowledge and Language-specific Knowledge标题：通过学习和结合一般语音知识和特定语言知识进行低资源语言的唇读链接：https://arxiv.org/abs/2308.09311

作者：Minsu Kim,Jeong Hun Yeo,Jeongsoo Choi,Yong Man Ro机构：School of Electrical Engineering, KAIST, South Korea备注：Accepted at ICCV 2023

【7】 Robust Audio Anti-Spoofing with Fusion-Reconstruction Learning on Multi-Order Spectrograms标题：基于多阶谱图融合重构学习的稳健音频反欺骗链接：https://arxiv.org/abs/2308.09302

作者：Penghui Wen,Kun Hu,Wenxi Yue,Sen Zhang,Wanlei Zhou,Zhiyong Wang机构：School of Computer Science, The University of Sydney, NSW, Australia, City University of Macau, Macao SAR, China

【8】 V2A-Mapper: A Lightweight Solution for Vision-to-Audio Generation by Connecting Foundation Models标题：V2a-Mapper：一种连接基础模型的轻量级视听生成解决方案链接：https://arxiv.org/abs/2308.09300

作者：Heng Wang,Jianbo Ma,Santiago Pascual,Richard Cartwright,Weidong Cai机构：Dolby Laboratories, University of Sydney备注：13 pages, 10 figures. Code, demo, and samples: this https URL

【9】 Bridging High-Quality Audio and Video via Language for Sound Effects Retrieval from Visual Queries标题：从视觉查询中检索音效的语言为高质量的音频和视频架起桥梁链接：https://arxiv.org/abs/2308.09089

作者：Julia Wilkins,Justin Salamon,Magdalena Fuentes,Juan Pablo Bello,Oriol Nieto机构：Music and Audio Research Laboratory, New York University, NY, USA, Adobe Research, San Francisco, CA, USA备注：WASPAA 2023. Project page: this https URL 4 pages, 2 figures, 2 tables

【10】 Long-frame-shift Neural Speech Phase Prediction with Spectral Continuity Enhancement and Interpolation Error Compensation标题：谱连续性增强和内插误差补偿的长帧移神经语音相位预测链接：https://arxiv.org/abs/2308.08850

作者：Yang Ai,Ye-Xin Lu,Zhen-Hua Ling机构：two-stage methods [ 1 1][ 1 2][ 1 3]This work was supported by the Fundamental Research Funds for theCentral Universities under Grant WK 2 1000000 3 3, Universityof Science and Technology of China备注：Published at IEEE Signal Processing Letters

【11】 Decoding Emotions: A comprehensive Multilingual Study of Speech Models for Speech Emotion Recognition标题：情绪解码：用于语音情绪识别的语音模型的综合多语言研究链接：https://arxiv.org/abs/2308.08713

作者：Anant Singh,Akshat Gupta机构：New York University, New York, USA, JP Morgan Chase

【12】 AffectEcho: Speaker Independent and Language-Agnostic Emotion and Affect Transfer for Speech Synthesis标题：AffectEcho：语音合成中的非特定人和语言不可知的情绪和情感传递链接：https://arxiv.org/abs/2308.08577

作者：Hrishikesh Viswanath,Aneesh Bhattacharya,Pascal Jutras-Dubé,Prerit Gupta,Mridu Prashanth,Yashvardhan Khaitan,Aniket Bera机构：Department of Computer Science, Purdue University, West Lafayette, IN, USA

【13】 Generative Machine Listener标题：产生式机器监听器链接：https://arxiv.org/abs/2308.09493

作者：Guanxin Jiang,Lars Villemoes,Arijit Biswas机构：Sound Tech Research, Dolby Germany GmbH, Dolby Sweden AB备注：Accepted to 155th Audio Engineering Society (AES) Convention, New York, NY, USA, October 2023

【14】 Multi-Task Pseudo-Label Learning for Non-Intrusive Speech Quality Assessment Model标题：用于非侵入式语音质量评估模型的多任务伪标签学习链接：https://arxiv.org/abs/2308.09262

作者：Ryandhimas E. Zezario,Bo-Ren Brian Bai,Chiou-Shann Fuh,Hsin-Min Wang,Yu Tsao机构：∗ Department of Computer Science and Information Engineering, National Taiwan University, Taiwan, † Research Center for Information Technology Innovation, Academia Sinica, Taiwan, ‡ Fortemedia, Taiwan, § Institute of Information Science, Academia Sinica, Taiwan

【15】 Refining a Deep Learning-based Formant Tracker using Linear Prediction Methods标题：用线性预测方法改进基于深度学习的共振峰跟踪器链接：https://arxiv.org/abs/2308.09051

作者：Paavo Alku,Sudarsana Reddy Kadiri,Dhananjaya Gowda机构：Department of Information and Communications Engineering, Aalto University, Finland, Samsung Research, South Korea备注：Computer Speech and Language, Vol. 81, Article 101515, June 2023

【16】 Severity Classification of Parkinson's Disease from Speech using Single Frequency Filtering-based Features标题：基于单频滤波的语音帕金森病严重程度分类链接：https://arxiv.org/abs/2308.09042

作者：Sudarsana Reddy Kadiri,Manila Kodali,Paavo Alku机构：Department of Information and Communications Engineering, Aalto University, Finland.备注：Accepted by INTERSPEECH 2023

【17】 Home monitoring for frailty detection through sound and speaker diarization analysis标题：通过声音和扬声器二值化分析进行脆弱性检测的家庭监控链接：https://arxiv.org/abs/2308.08985

作者：Yannis Tevissen,Dan Istrate,Vincent Zalc,Jérôme Boudy,Gérard Chollet,Frédéric Petitpont,Sami Boutamine机构：Université de technologie de Compiègne, CNRS, Biomécanique et Bioingénierie, Centre de Recherche Royallieu – CS , -, Compiègne Cedex, France, SAMOVAR, Télécom SudParis, Institut Polytechnique de Paris, Palaiseau, France备注：JETSAN, Jun 2023, Aubervilliers & Paris, France

【18】 Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancement标题：用于高质量语音增强的并行显式幅度谱和相位谱估计链接：https://arxiv.org/abs/2308.08926

作者：Ye-Xin Lu,Yang Ai,Zhen-Hua Ling备注：Submmited to IEEE Transactions on Audio, Speech and Language Processing

【19】 META-SELD: Meta-Learning for Fast Adaptation to the new environment in Sound Event Localization and Detection标题：元学习：声事件定位与检测中快速适应新环境的元学习链接：https://arxiv.org/abs/2308.08847

作者：Jinbo Hu,Yin Cao,Ming Wu,Feiran Yang,Ziying Yu,Wenwu Wang,Mark D. Plumbley,Jun Yang机构：Key Laboratory of Noise and Vibration Research, Institute of Acoustics, University of Chinese Academy of Sciences, Beijing, China, Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey, UK备注：Submitted to DCASE 2023 Workshop

【20】 Graph Neural Network Backend for Speaker Recognition标题：用于说话人识别的图神经网络后端链接：https://arxiv.org/abs/2308.08767

作者：Liang He,Ruida Li,Mengqi Niu机构：Ruida Li and Mengqi Niu are with School of Information Scienceand Engineering, Xinjiang University, Liang He isalso with the Department of Electronic Engineering, Tsinghua University

【21】 The DKU-MSXF Speaker Verification System for the VoxCeleb Speaker Recognition Challenge 2023标题：用于VoxCeleb说话人识别挑战赛2023的DKU-MSXF说话人确认系统链接：https://arxiv.org/abs/2308.08766

作者：Ze Li,Yuke Lin,Xiaoyi Qin,Ning Jiang,Guoqing Zhao,Ming Li机构：Data Science Research Center, Duke Kunshan University, Kunshan, China, Mashang Consumer Finance Co., Ltd, China备注：arXiv admin note: text overlap with arXiv:2210.05092

eess.AS音频处理

【1】 Generative Machine Listener标题：产生式机器监听器链接：https://arxiv.org/abs/2308.09493

【2】 Multi-Task Pseudo-Label Learning for Non-Intrusive Speech Quality Assessment Model标题：用于非侵入式语音质量评估模型的多任务伪标签学习链接：https://arxiv.org/abs/2308.09262

【3】 Refining a Deep Learning-based Formant Tracker using Linear Prediction Methods标题：用线性预测方法改进基于深度学习的共振峰跟踪器链接：https://arxiv.org/abs/2308.09051

【4】 Severity Classification of Parkinson's Disease from Speech using Single Frequency Filtering-based Features标题：基于单频滤波的语音帕金森病严重程度分类链接：https://arxiv.org/abs/2308.09042

作者：Sudarsana Reddy Kadiri,Manila Kodali,Paavo Alku机构：Department of Information and Communications Engineering, Aalto University, Finland.备注：Accepted by INTERSPEECH 2023

【5】 Home monitoring for frailty detection through sound and speaker diarization analysis标题：通过声音和扬声器二值化分析进行脆弱性检测的家庭监控链接：https://arxiv.org/abs/2308.08985

【6】 Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancement标题：用于高质量语音增强的并行显式幅度谱和相位谱估计链接：https://arxiv.org/abs/2308.08926

作者：Ye-Xin Lu,Yang Ai,Zhen-Hua Ling备注：Submmited to IEEE Transactions on Audio, Speech and Language Processing

【7】 META-SELD: Meta-Learning for Fast Adaptation to the new environment in Sound Event Localization and Detection标题：元学习：声事件定位与检测中快速适应新环境的元学习链接：https://arxiv.org/abs/2308.08847

【8】 Graph Neural Network Backend for Speaker Recognition标题：用于说话人识别的图神经网络后端链接：https://arxiv.org/abs/2308.08767

【9】 The DKU-MSXF Speaker Verification System for the VoxCeleb Speaker Recognition Challenge 2023标题：用于VoxCeleb说话人识别挑战赛2023的DKU-MSXF说话人确认系统链接：https://arxiv.org/abs/2308.08766

【10】 Audiovisual Moments in Time: A Large-Scale Annotated Dataset of Audiovisual Actions标题：时间中的视听时刻：视听动作的大规模注释数据集链接：https://arxiv.org/abs/2308.09685

【11】 Compensating Removed Frequency Components: Thwarting Voice Spectrum Reduction Attacks标题：补偿去除的频率分量：挫败语音频谱降低攻击链接：https://arxiv.org/abs/2308.09546

作者：Shu Wang,Kun Sun,Qi Li机构：George Mason University, Tsinghua University备注：Accepted by 2024 Network and Distributed System Security Symposium (NDSS'24)

【12】 Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learning标题：Space LibriSpeech：一种用于空间音频学习的扩充数据集链接：https://arxiv.org/abs/2308.09514

【13】 Exploring Sampling Techniques for Generating Melodies with a Transformer Language Model标题：探索用Transformer语言模型生成旋律的采样技术链接：https://arxiv.org/abs/2308.09454

【14】 TrOMR:Transformer-Based Polyphonic Optical Music Recognition标题：TrOMR：基于Transformer的复调光学音乐识别链接：https://arxiv.org/abs/2308.09370

作者：Yixuan Li,Huaping Liu,Qiang Jin,Miaomiao Cai,Peng Li机构：NetEase Cloud Music, Hangzhou, China备注：None

【15】 Lip Reading for Low-resource Languages by Learning and Combining General Speech Knowledge and Language-specific Knowledge标题：通过学习和结合一般语音知识和特定语言知识进行低资源语言的唇读链接：https://arxiv.org/abs/2308.09311

作者：Minsu Kim,Jeong Hun Yeo,Jeongsoo Choi,Yong Man Ro机构：School of Electrical Engineering, KAIST, South Korea备注：Accepted at ICCV 2023

【16】 Robust Audio Anti-Spoofing with Fusion-Reconstruction Learning on Multi-Order Spectrograms标题：基于多阶谱图融合重构学习的稳健音频反欺骗链接：https://arxiv.org/abs/2308.09302

作者：Penghui Wen,Kun Hu,Wenxi Yue,Sen Zhang,Wanlei Zhou,Zhiyong Wang机构：School of Computer Science, The University of Sydney, NSW, Australia, City University of Macau, Macao SAR, China

【17】 V2A-Mapper: A Lightweight Solution for Vision-to-Audio Generation by Connecting Foundation Models标题：V2a-Mapper：一种连接基础模型的轻量级视听生成解决方案链接：https://arxiv.org/abs/2308.09300

作者：Heng Wang,Jianbo Ma,Santiago Pascual,Richard Cartwright,Weidong Cai机构：Dolby Laboratories, University of Sydney备注：13 pages, 10 figures. Code, demo, and samples: this https URL

【18】 Bridging High-Quality Audio and Video via Language for Sound Effects Retrieval from Visual Queries标题：从视觉查询中检索音效的语言为高质量的音频和视频架起桥梁链接：https://arxiv.org/abs/2308.09089

【19】 Long-frame-shift Neural Speech Phase Prediction with Spectral Continuity Enhancement and Interpolation Error Compensation标题：谱连续性增强和内插误差补偿的长帧移神经语音相位预测链接：https://arxiv.org/abs/2308.08850

【20】 Decoding Emotions: A comprehensive Multilingual Study of Speech Models for Speech Emotion Recognition标题：情绪解码：用于语音情绪识别的语音模型的综合多语言研究链接：https://arxiv.org/abs/2308.08713

作者：Anant Singh,Akshat Gupta机构：New York University, New York, USA, JP Morgan Chase

【21】 AffectEcho: Speaker Independent and Language-Agnostic Emotion and Affect Transfer for Speech Synthesis标题：AffectEcho：语音合成中的非特定人和语言不可知的情绪和情感传递链接：https://arxiv.org/abs/2308.08577

机器翻译，仅供参考

语音/音频处理学术速递[8.21]

评论 0