今日论文合集：eess.AS音频处理21篇，cs.SD语音15篇。【1】 Disentangling Voice and Content with Self-Supervision for Speaker Recognition标题：说话人识别中的语音与内容自监督解缠作者：Tianchi Liu,Kong Aik Lee,Qiongqiong Wang,Haizhou Li机构：Institute for Infocomm Research (I,R), Agency for Science, Technology and Research (A⋆STAR), Singapore, Dept. of Electrical and Computer Engineering, National University of Singapore, Singapore备注：Accepted to NeurIPS 2023 (main track)链接：http://arxiv.org/pdf/2310.01128v1【2】 Scaling Up Music Information Retrieval Training with Semi-Supervised Learning标题：基于半监督学习的音乐信息提取训练作者：Yun-Ning Hung,Ju-Chiang Wang,Minz Won,Duc Le机构：SAMI, ByteDance, San Jose, CA, USA链接：http://arxiv.org/pdf/2310.01353v1【3】 A Fused Deep Denoising Sound Coding Strategy for Bilateral Cochlear Implants标题：一种融合深度去噪的双侧人工耳蜗声编码策略作者：Tom Gajecki,Waldo Nogueira机构：and WN are with the Department of Otolaryngology, the MedicalUniversity of Hannover and Cluster of Excellence链接：http://arxiv.org/pdf/2310.01122v1【4】 Wavelet Scattering Transform for Improving Generalization in Low-Resourced Spoken Language Identification标题：小波散射变换提高低资源口语识别泛化能力作者：Spandan Dey,Premjeet Singh,Goutam Saha机构：Department of Electronics & Electrical Communication Engineering, Indian Institute of Technology Kharagpur, West Bengal, India备注：Accepted in INTERSPEECH 2023链接：http://arxiv.org/pdf/2310.00602v1【5】 Mechatronic Generation of Datasets for Acoustics Research标题：声学研究数据集的机电一体化生成作者：Austin Lu,Ethaniel Moore,Arya Nallanthighall,Kanad Sarkar,Manan Mittal,Ryan M. Corey,Paris Smaragdis,Andrew Singer机构：University of Illinois Urbana-Champaign备注：5 pages, 5 figures, IWAENC 2022链接：http://arxiv.org/pdf/2310.00587v1【6】 Time-Variant Overlap-Add in Partitions标题：时变重叠-添加分区作者：Hagen Jaeger,Uwe Simmer,Jörg Bitzer,Matthias Blau链接：http://arxiv.org/pdf/2310.00319v1【7】 A Novel U-Net Architecture for Denoising of Real-world Noise Corrupted Phonocardiogram Signal标题：一种用于心音信号去噪的新U-网结构作者：Ayan Mukherjee,Rohan Banerjee,Avik Ghose机构：TCS Research, Tata Consultancy Services, Kolkata, West Bengal, India链接：http://arxiv.org/pdf/2310.00216v1【8】 DiffAR: Denoising Diffusion Autoregressive Model for Raw Speech Waveform Generation标题：DiffAR：用于原始语音波形生成的去噪扩散自回归模型作者：Roi Benita,Michael Elad,Joseph Keshet机构：Department of Electrical and Computer Engineering, Department of Computer Science, Technion - Israel Institute of Technology, Haifa, Israel链接：http://arxiv.org/pdf/2310.01381v1【9】 Towards human-like spoken dialogue generation between AI agents from written dialogue标题：从书面对话到人工智能主体之间的类人类口语对话生成作者：Kentaro Mitsui,Yukiya Hono,Kei Sawada机构：rinna Co., Ltd., Tokyo, Japan备注：18 pages, 8 figures, 9 tables, audio samples: this https URL链接：http://arxiv.org/pdf/2310.01088v1【10】 uSee: Unified Speech Enhancement and Editing with Conditional Diffusion Models标题：用途：使用条件扩散模型进行统一的语音增强和编辑作者：Muqiao Yang,Chunlei Zhang,Yong Xu,Zhongweiyang Xu,Heming Wang,Bhiksha Raj,Dong Yu机构：Carnegie Mellon University, Tencent AI Lab, University of Illinois Urbana-Champaign, The Ohio State University, MBZUAI链接：http://arxiv.org/pdf/2310.00900v1【11】 F0 analysis of Ghanaian pop singing reveals progressive alignment with equal temperament over the past three decades: a case study标题：对加纳流行歌唱的F0分析显示，在过去的30年里，以同样的气质逐步趋同：一个案例研究作者：Iran R. Roman,Daniel Faronbi,Isabelle Burger-Weiser,Leila Adu-Gilmore机构：New York University备注：Pages 27-33链接：http://arxiv.org/pdf/2310.00870v1【12】 Evaluating Speech Synthesis by Training Recognizers on Synthetic Speech标题：通过对合成语音训练识别器来评价语音合成作者：Dareen Alharthi,Roshan Sharma,Hira Dhamyal,Soumi Maiti,Bhiksha Raj,Rita Singh机构：Carnegie Mellon University,Mohammed bin Zayed University of AI链接：http://arxiv.org/pdf/2310.00706v1【13】 UniAudio: An Audio Foundation Model Toward Universal Audio Generation标题：UniAudio：面向通用音频生成的音频基础模型作者：Dongchao Yang,Jinchuan Tian,Xu Tan,Rongjie Huang,Songxiang Liu,Xuankai Chang,Jiatong Shi,Sheng Zhao,Jiang Bian,Xixin Wu,Zhou Zhao,Helen Meng机构：The Chinese University of Hong Kong, Carnegie Mellon University, Microsoft Research Asia, Zhejiang University链接：http://arxiv.org/pdf/2310.00704v1【14】 Pianist Identification Using Convolutional Neural Networks标题：基于卷积神经网络的钢琴演奏者识别作者：Jingjing Tang,Geraint Wiggins,Gyorgy Fazekas机构：Center for Digital Music, Computational Creativity Lab, Vrije Universiteit Brussel, Belgium &, Queen Mary University of London, UK, rd Gy¨orgy Fazekas备注：6 pages, 3 figures, accepted by the 4th International Symposium on the Internet of Sounds, IS2 2023链接：http://arxiv.org/pdf/2310.00699v1【15】 Music- and Lyrics-driven Dance Synthesis标题：音乐和歌词驱动的舞蹈合成作者：Wenjie Yin,Qingyuan Yao,Yi Yu,Hang Yin,Danica Kragic,Mårten Björkman机构：KTH Royal Institute of Technology, Sorbonne University, National Institute of Informatics, University of Copenhagen链接：http://arxiv.org/pdf/2310.00455v1【16】 Active Learning Based Fine-Tuning Framework for Speech Emotion Recognition标题：基于主动学习的语音情感识别微调框架作者：Dongyuan Li,Yusong Wang,Kotaro Funakoshi,Manabu Okumura机构：Tokyo Institute of Technology.链接：http://arxiv.org/pdf/2310.00283v1【17】 SLM: Bridge the thin gap between speech and text foundation models标题：SLM：弥合语音和文本基础模型之间的细小差距作者：Mingqiu Wang,Wei Han,Izhak Shafran,Zelin Wu,Chung-Cheng Chiu,Yuan Cao,Yongqiang Wang,Nanxin Chen,Yu Zhang,Hagen Soltau,Paul Rubenstein,Lukas Zilka,Dian Yu,Zhong Meng,Golan Pundak,Nikhil Siddhartha,Johan Schalkwyk,Yonghui Wu机构：Google Deepmind链接：http://arxiv.org/pdf/2310.00230v1【18】 Contextual Biasing with the Knuth-Morris-Pratt Matching Algorithm标题：Knuth-Morris-Pratt匹配算法的语境偏向作者：Weiran Wang,Zelin Wu,Diamantino Caseiro,Tsendsuren Munkhdalai,Khe Chai Sim,Pat Rondon,Golan Pundak,Gan Song,Rohit Prabhavalkar,Zhong Meng,Ding Zhao,Tara Sainath,Pedro Moreno Mengibar机构：Yanzhang He, Google链接：http://arxiv.org/pdf/2310.00178v1【19】 The Gift of Feedback: Improving ASR Model Quality by Learning from User Corrections through Federated Learning标题：反馈的礼物：通过联合学习从用户校正中学习来提高ASR模型质量作者：Lillian Zhou,Yuxin Ding,Mingqing Chen,Harry Zhang,Rohit Prabhavalkar,Dhruv Guliani,Giovanni Motta,Rajiv Mathews机构：Google LLC, Mountain View, CA, U.S.A.备注：Accepted to IEEE ASRU 2023链接：http://arxiv.org/pdf/2310.00141v1【20】 GASS: Generalizing Audio Source Separation with Large-scale Data标题：GASS：推广大规模数据的音频源分离作者：Jordi Pons,Xiaoyu Liu,Santiago Pascual,Joan Serrà机构：Joan Serra, Dolby Laboratories链接：http://arxiv.org/pdf/2310.00140v1【21】 Fewer-token Neural Speech Codec with Time-invariant Codes标题：基于时不变码的少令牌神经语音编解码器作者：Yong Ren,Tao Wang,Jiangyan Yi,Le Xu,Jianhua Tao,Chuyuan Zhang,Junzuo Zhou机构：Institute of Automation, Chinese Academy of Sciences, China, University of Chinese Academy of Sciences, China, Department of Automation, Tsinghua University, China备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2310.00014v1【1】 DiffAR: Denoising Diffusion Autoregressive Model for Raw Speech Waveform Generation标题：DiffAR：用于原始语音波形生成的去噪扩散自回归模型作者：Roi Benita,Michael Elad,Joseph Keshet机构：Department of Electrical and Computer Engineering, Department of Computer Science, Technion - Israel Institute of Technology, Haifa, Israel链接：http://arxiv.org/pdf/2310.01381v1
【2】 Towards human-like spoken dialogue generation between AI agents from written dialogue标题：从书面对话到人工智能主体之间的类人类口语对话生成作者：Kentaro Mitsui,Yukiya Hono,Kei Sawada机构：rinna Co., Ltd., Tokyo, Japan备注：18 pages, 8 figures, 9 tables, audio samples: this https URL链接：http://arxiv.org/pdf/2310.01088v1
【3】 uSee: Unified Speech Enhancement and Editing with Conditional Diffusion Models标题：用途：使用条件扩散模型进行统一的语音增强和编辑作者：Muqiao Yang,Chunlei Zhang,Yong Xu,Zhongweiyang Xu,Heming Wang,Bhiksha Raj,Dong Yu机构：Carnegie Mellon University, Tencent AI Lab, University of Illinois Urbana-Champaign, The Ohio State University, MBZUAI链接：http://arxiv.org/pdf/2310.00900v1【4】 F0 analysis of Ghanaian pop singing reveals progressive alignment with equal temperament over the past three decades: a case study标题：对加纳流行歌唱的F0分析显示，在过去的30年里，以同样的气质逐步趋同：一个案例研究作者：Iran R. Roman,Daniel Faronbi,Isabelle Burger-Weiser,Leila Adu-Gilmore机构：New York University备注：Pages 27-33链接：http://arxiv.org/pdf/2310.00870v1【5】 Evaluating Speech Synthesis by Training Recognizers on Synthetic Speech标题：通过对合成语音训练识别器来评价语音合成作者：Dareen Alharthi,Roshan Sharma,Hira Dhamyal,Soumi Maiti,Bhiksha Raj,Rita Singh机构：Carnegie Mellon University,Mohammed bin Zayed University of AI链接：http://arxiv.org/pdf/2310.00706v1【6】 UniAudio: An Audio Foundation Model Toward Universal Audio Generation标题：UniAudio：面向通用音频生成的音频基础模型作者：Dongchao Yang,Jinchuan Tian,Xu Tan,Rongjie Huang,Songxiang Liu,Xuankai Chang,Jiatong Shi,Sheng Zhao,Jiang Bian,Xixin Wu,Zhou Zhao,Helen Meng机构：The Chinese University of Hong Kong, Carnegie Mellon University, Microsoft Research Asia, Zhejiang University链接：http://arxiv.org/pdf/2310.00704v1【7】 Pianist Identification Using Convolutional Neural Networks标题：基于卷积神经网络的钢琴演奏者识别作者：Jingjing Tang,Geraint Wiggins,Gyorgy Fazekas机构：Center for Digital Music, Computational Creativity Lab, Vrije Universiteit Brussel, Belgium &, Queen Mary University of London, UK, rd Gy¨orgy Fazekas备注：6 pages, 3 figures, accepted by the 4th International Symposium on the Internet of Sounds, IS2 2023链接：http://arxiv.org/pdf/2310.00699v1【8】 Music- and Lyrics-driven Dance Synthesis标题：音乐和歌词驱动的舞蹈合成作者：Wenjie Yin,Qingyuan Yao,Yi Yu,Hang Yin,Danica Kragic,Mårten Björkman机构：KTH Royal Institute of Technology, Sorbonne University, National Institute of Informatics, University of Copenhagen链接：http://arxiv.org/pdf/2310.00455v1【9】 Active Learning Based Fine-Tuning Framework for Speech Emotion Recognition标题：基于主动学习的语音情感识别微调框架作者：Dongyuan Li,Yusong Wang,Kotaro Funakoshi,Manabu Okumura机构：Tokyo Institute of Technology.链接：http://arxiv.org/pdf/2310.00283v1【10】 SLM: Bridge the thin gap between speech and text foundation models标题：SLM：弥合语音和文本基础模型之间的细小差距作者：Mingqiu Wang,Wei Han,Izhak Shafran,Zelin Wu,Chung-Cheng Chiu,Yuan Cao,Yongqiang Wang,Nanxin Chen,Yu Zhang,Hagen Soltau,Paul Rubenstein,Lukas Zilka,Dian Yu,Zhong Meng,Golan Pundak,Nikhil Siddhartha,Johan Schalkwyk,Yonghui Wu机构：Google Deepmind链接：http://arxiv.org/pdf/2310.00230v1【11】 GASS: Generalizing Audio Source Separation with Large-scale Data标题：GASS：推广大规模数据的音频源分离作者：Jordi Pons,Xiaoyu Liu,Santiago Pascual,Joan Serrà机构：Joan Serra, Dolby Laboratories链接：http://arxiv.org/pdf/2310.00140v1【12】 Fewer-token Neural Speech Codec with Time-invariant Codes标题：基于时不变码的少令牌神经语音编解码器作者：Yong Ren,Tao Wang,Jiangyan Yi,Le Xu,Jianhua Tao,Chuyuan Zhang,Junzuo Zhou机构：Institute of Automation, Chinese Academy of Sciences, China, University of Chinese Academy of Sciences, China, Department of Automation, Tsinghua University, China备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2310.00014v1【13】 Scaling Up Music Information Retrieval Training with Semi-Supervised Learning标题：基于半监督学习的音乐信息提取训练作者：Yun-Ning Hung,Ju-Chiang Wang,Minz Won,Duc Le机构：SAMI, ByteDance, San Jose, CA, USA链接：http://arxiv.org/pdf/2310.01353v1【14】 Time-Variant Overlap-Add in Partitions标题：时变重叠-添加分区作者：Hagen Jaeger,Uwe Simmer,Jörg Bitzer,Matthias Blau链接：http://arxiv.org/pdf/2310.00319v1【15】 A Novel U-Net Architecture for Denoising of Real-world Noise Corrupted Phonocardiogram Signal标题：一种用于心音信号去噪的新U-网结构作者：Ayan Mukherjee,Rohan Banerjee,Avik Ghose机构：TCS Research, Tata Consultancy Services, Kolkata, West Bengal, India链接：http://arxiv.org/pdf/2310.00216v1
每日语音速递