今日论文合集：cs.SD语音18篇，eess.AS音频处理18篇。
cs.SD语音
【1】 Audio-Visual Speaker Verification via Joint Cross-Attention标题：基于联合交叉注意的视听说话人确认作者：R. Gnana Praveen,Jahangir Alam机构：Computer Research Institute of Montreal, Montreal, Quebec H,N ,M, Canada链接：http://arxiv.org/pdf/2309.16569v1
【2】 Efficient Supervised Training of Audio Transformers for Music Representation Learning标题：用于音乐表征学习的音频转换器的高效有监督训练作者：Pablo Alonso-Jiménez,Xavier Serra,Dmitry Bogdanov机构：Music Technology Group, Universitat Pompeu Fabra, Barcelona备注：Accepted at the 2023 International Society for Music Information Retrieval Conference (ISMIR'23)链接：http://arxiv.org/pdf/2309.16418v1
【3】 Bringing the Discussion of Minima Sharpness to the Audio Domain: a Filter-Normalised Evaluation for Acoustic Scene Classification标题：将微小图像清晰度的讨论带入音频域：一种用于声场分类的滤波归一化评价作者：Manuel Milling,Andreas Triantafyllopoulos,Iosif Tsangko,Simon David Noel Rampp,Björn Wolfgang Schuller机构：Chair of Embedded Intelligence for Health Care and Wellbeing, University of Augsburg, Germany, GLAM – Group on Language, Audio, & Music, Imperial College London, UK备注：This work has been submitted to the IEEE for possible publication链接：http://arxiv.org/pdf/2309.16369v1
【4】 Audio Visual Speaker Localization from EgoCentric Views标题：以自我为中心的视听说话人本土化作者：Jinzheng Zhao,Yong Xu,Xinyuan Qian,Wenwu Wang机构：University of Science and Technology Beijing链接：http://arxiv.org/pdf/2309.16308v1【5】 Predicting performance difficulty from piano sheet music images标题：从钢琴乐谱图像预测演奏难度作者：Pedro Ramoneda,Jose J. Valero-Mas,Dasaem Jeong,Xavier Serra机构：Music Technology Group, Universitat Pompeu Fabra, Barcelona, MALer Lab, Sogang University, Seoul链接：http://arxiv.org/pdf/2309.16287v1【6】 NOMAD: Unsupervised Learning of Perceptual Embeddings for Speech Enhancement and Non-matching Reference Audio Quality Assessment标题：Nomad：用于语音增强和非匹配参考音频质量评估的感知嵌入的无监督学习作者：Alessandro Ragano,Jan Skoglund,Andrew Hines机构：School of Computer Science, University College Dublin, Ireland,Google LLC, San Francisco, USA备注：5 pages链接：http://arxiv.org/pdf/2309.16284v1【7】 Semantic Proximity Alignment: Towards Human Perception-consistent Audio Tagging by Aligning with Label Text Description标题：语义接近对齐：通过与标签文本描述对齐来实现人类感知一致的音频标记作者：Wuyang Liu,Yanzhen Ren机构：Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University备注：5 pages, 3 figures. Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16265v1【8】 LAE-ST-MoE: Boosted Language-Aware Encoder Using Speech Translation Auxiliary Task for E2E Code-switching ASR标题：LAE-ST-MOE：基于语音翻译辅助任务的增强型语言感知编码器作者：Guodong Ma,Wenxuan Wang,Yuke Li,Yuting Yang,Binbin Du,Haoran Fu机构：†NetEase Yidun AI Lab, Hangzhou, China, ‡Department of Civil Engineering, Zhejiang University备注：Accepted to IEEE ASRU 2023链接：http://arxiv.org/pdf/2309.16178v1【9】 Neural Acoustic Context Field: Rendering Realistic Room Impulse Response With Neural Fields标题：神经声学上下文场：用神经场再现逼真的房间脉冲响应作者：Susan Liang,Chao Huang,Yapeng Tian,Anurag Kumar,Chenliang Xu机构：University of Rochester链接：http://arxiv.org/pdf/2309.15977v1【10】 Unsupervised Pre-Training for Vietnamese Automatic Speech Recognition in the HYKIST Project标题：HYKIST项目中越南语自动语音识别的无监督预训练作者：Khai Le-DucJournal-ref：FH Aachen University of Applied Sciences (2023)链接：http://arxiv.org/pdf/2309.15869v1【11】 Meeting Recognition with Continuous Speech Separation and Transcription-Supported Diarization标题：基于连续语音分离和支持转录的二值化的会议识别作者：Thilo von Neumann,Christoph Boeddeker,Tobias Cord-Landwehr,Marc Delcroix,Reinhold Haeb-Umbach机构：Paderborn University, NTT corporation, Japan备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16482v1
【12】 PP-MeT: a Real-world Personalized Prompt based Meeting Transcription System标题：PP-MET：一个基于真实世界个性化提示的会议转录系统作者：Xiang Lyu,Yuhang Cao,Qing Wang,Jingjing Yin,Yuguang Yang,Pengpeng Zou,Yanni Hu,Heng Lu机构：Ximalaya Inc., Shanghai, China链接：http://arxiv.org/pdf/2309.16247v1【13】 Hierarchical Cross-Modality Knowledge Transfer with Sinkhorn Attention for CTC-based ASR标题：基于CTC的ASR中基于辛克霍恩注意力的层次化跨通道知识传递作者：Xugang Lu,Peng Shen,Yu Tsao,Hisashi Kawai机构：. National Institute of Information and Communications Technology, Japan, . Research Center for Information Technology Innovation, Academic Sinica, Taiwan备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16093v1【14】 Does Single-channel Speech Enhancement Improve Keyword Spotting Accuracy? A Case Study标题：单通道语音增强是否提高了关键词识别的准确率？案例研究作者：Avamarie Brueggeman,Takuya Higuchi,Masood Delfarah,Stephen Shum,Vineet Garg机构：University of Texas at Dallas, USA, Apple, USA链接：http://arxiv.org/pdf/2309.16060v1【15】 Neural Network Augmented Kalman Filter for Robust Acoustic Howling Suppression标题：神经网络增广卡尔曼滤波在抗声咆哮中的应用作者：Yixuan Zhang,Hao Zhang,Meng Yu,Dong Yu机构：The Ohio State University, Columbus, OH, USA, Tencent AI Lab, Bellevue, WA, USA备注：Paper in submission链接：http://arxiv.org/pdf/2309.16049v1【16】 Advancing Acoustic Howling Suppression through Recursive Training of Neural Networks标题：基于神经网络递归训练的声学咆哮抑制作者：Hao Zhang,Yixuan Zhang,Meng Yu,Dong Yu机构：Tencent AI Lab, Bellevue, WA, USA, The Ohio State University, Columbus, OH, USA备注：Paper in submission链接：http://arxiv.org/pdf/2309.16048v1【17】 Multichannel Voice Trigger Detection Based on Transform-average-concatenate标题：基于变换平均级联的多通道语音触发检测作者：Takuya Higuchi,Avamarie Brueggeman,Masood Delfarah,Stephen Shum机构：Apple, USA, The University of Texas at Dallas, USA链接：http://arxiv.org/pdf/2309.16036v1【18】 Exploring Self-Supervised Contrastive Learning of Spatial Sound Event Representation标题：空间声事件表征的自监督对比学习探索作者：Xilin Jiang,Cong Han,Yinghao Aaron Li,Nima Mesgarani机构：Department of Electrical Engineering, Columbia University, USA链接：http://arxiv.org/pdf/2309.15938v1
eess.AS音频处理
【1】 Meeting Recognition with Continuous Speech Separation and Transcription-Supported Diarization标题：基于连续语音分离和支持转录的二值化的会议识别作者：Thilo von Neumann,Christoph Boeddeker,Tobias Cord-Landwehr,Marc Delcroix,Reinhold Haeb-Umbach机构：Paderborn University, NTT corporation, Japan备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16482v1
【2】 PP-MeT: a Real-world Personalized Prompt based Meeting Transcription System标题：PP-MET：一个基于真实世界个性化提示的会议转录系统作者：Xiang Lyu,Yuhang Cao,Qing Wang,Jingjing Yin,Yuguang Yang,Pengpeng Zou,Yanni Hu,Heng Lu机构：Ximalaya Inc., Shanghai, China链接：http://arxiv.org/pdf/2309.16247v1
【3】 Hierarchical Cross-Modality Knowledge Transfer with Sinkhorn Attention for CTC-based ASR标题：基于CTC的ASR中基于辛克霍恩注意力的层次化跨通道知识传递作者：Xugang Lu,Peng Shen,Yu Tsao,Hisashi Kawai机构：. National Institute of Information and Communications Technology, Japan, . Research Center for Information Technology Innovation, Academic Sinica, Taiwan备注：Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16093v1【4】 Does Single-channel Speech Enhancement Improve Keyword Spotting Accuracy? A Case Study标题：单通道语音增强是否提高了关键词识别的准确率？案例研究作者：Avamarie Brueggeman,Takuya Higuchi,Masood Delfarah,Stephen Shum,Vineet Garg机构：University of Texas at Dallas, USA, Apple, USA链接：http://arxiv.org/pdf/2309.16060v1
【5】 Neural Network Augmented Kalman Filter for Robust Acoustic Howling Suppression标题：神经网络增广卡尔曼滤波在抗声咆哮中的应用作者：Yixuan Zhang,Hao Zhang,Meng Yu,Dong Yu机构：The Ohio State University, Columbus, OH, USA, Tencent AI Lab, Bellevue, WA, USA备注：Paper in submission链接：http://arxiv.org/pdf/2309.16049v1【6】 Advancing Acoustic Howling Suppression through Recursive Training of Neural Networks标题：基于神经网络递归训练的声学咆哮抑制作者：Hao Zhang,Yixuan Zhang,Meng Yu,Dong Yu机构：Tencent AI Lab, Bellevue, WA, USA, The Ohio State University, Columbus, OH, USA备注：Paper in submission链接：http://arxiv.org/pdf/2309.16048v1【7】 Multichannel Voice Trigger Detection Based on Transform-average-concatenate标题：基于变换平均级联的多通道语音触发检测作者：Takuya Higuchi,Avamarie Brueggeman,Masood Delfarah,Stephen Shum机构：Apple, USA, The University of Texas at Dallas, USA链接：http://arxiv.org/pdf/2309.16036v1
【8】 Exploring Self-Supervised Contrastive Learning of Spatial Sound Event Representation标题：空间声事件表征的自监督对比学习探索作者：Xilin Jiang,Cong Han,Yinghao Aaron Li,Nima Mesgarani机构：Department of Electrical Engineering, Columbia University, USA链接：http://arxiv.org/pdf/2309.15938v1【9】 Audio-Visual Speaker Verification via Joint Cross-Attention标题：基于联合交叉注意的视听说话人确认作者：R. Gnana Praveen,Jahangir Alam机构：Computer Research Institute of Montreal, Montreal, Quebec H,N ,M, Canada链接：http://arxiv.org/pdf/2309.16569v1【10】 Efficient Supervised Training of Audio Transformers for Music Representation Learning标题：用于音乐表征学习的音频转换器的高效有监督训练作者：Pablo Alonso-Jiménez,Xavier Serra,Dmitry Bogdanov机构：Music Technology Group, Universitat Pompeu Fabra, Barcelona备注：Accepted at the 2023 International Society for Music Information Retrieval Conference (ISMIR'23)链接：http://arxiv.org/pdf/2309.16418v1【11】 Bringing the Discussion of Minima Sharpness to the Audio Domain: a Filter-Normalised Evaluation for Acoustic Scene Classification标题：将微小图像清晰度的讨论带入音频域：一种用于声场分类的滤波归一化评价作者：Manuel Milling,Andreas Triantafyllopoulos,Iosif Tsangko,Simon David Noel Rampp,Björn Wolfgang Schuller机构：Chair of Embedded Intelligence for Health Care and Wellbeing, University of Augsburg, Germany, GLAM – Group on Language, Audio, & Music, Imperial College London, UK备注：This work has been submitted to the IEEE for possible publication链接：http://arxiv.org/pdf/2309.16369v1【12】 Audio Visual Speaker Localization from EgoCentric Views标题：以自我为中心的视听说话人本土化作者：Jinzheng Zhao,Yong Xu,Xinyuan Qian,Wenwu Wang机构：University of Science and Technology Beijing链接：http://arxiv.org/pdf/2309.16308v1【13】 Predicting performance difficulty from piano sheet music images标题：从钢琴乐谱图像预测演奏难度作者：Pedro Ramoneda,Jose J. Valero-Mas,Dasaem Jeong,Xavier Serra机构：Music Technology Group, Universitat Pompeu Fabra, Barcelona, MALer Lab, Sogang University, Seoul链接：http://arxiv.org/pdf/2309.16287v1【14】 NOMAD: Unsupervised Learning of Perceptual Embeddings for Speech Enhancement and Non-matching Reference Audio Quality Assessment标题：Nomad：用于语音增强和非匹配参考音频质量评估的感知嵌入的无监督学习作者：Alessandro Ragano,Jan Skoglund,Andrew Hines机构：School of Computer Science, University College Dublin, Ireland,Google LLC, San Francisco, USA备注：5 pages链接：http://arxiv.org/pdf/2309.16284v1【15】 Semantic Proximity Alignment: Towards Human Perception-consistent Audio Tagging by Aligning with Label Text Description标题：语义接近对齐：通过与标签文本描述对齐来实现人类感知一致的音频标记作者：Wuyang Liu,Yanzhen Ren机构：Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University备注：5 pages, 3 figures. Submitted to ICASSP 2024链接：http://arxiv.org/pdf/2309.16265v1【16】 LAE-ST-MoE: Boosted Language-Aware Encoder Using Speech Translation Auxiliary Task for E2E Code-switching ASR标题：LAE-ST-MOE：基于语音翻译辅助任务的增强型语言感知编码器作者：Guodong Ma,Wenxuan Wang,Yuke Li,Yuting Yang,Binbin Du,Haoran Fu机构：†NetEase Yidun AI Lab, Hangzhou, China, ‡Department of Civil Engineering, Zhejiang University备注：Accepted to IEEE ASRU 2023链接：http://arxiv.org/pdf/2309.16178v1【17】 Neural Acoustic Context Field: Rendering Realistic Room Impulse Response With Neural Fields标题：神经声学上下文场：用神经场再现逼真的房间脉冲响应作者：Susan Liang,Chao Huang,Yapeng Tian,Anurag Kumar,Chenliang Xu机构：University of Rochester链接：http://arxiv.org/pdf/2309.15977v1【18】 Unsupervised Pre-Training for Vietnamese Automatic Speech Recognition in the HYKIST Project标题：HYKIST项目中越南语自动语音识别的无监督预训练作者：Khai Le-DucJournal-ref：FH Aachen University of Applied Sciences (2023)链接：http://arxiv.org/pdf/2309.15869v1
机器翻译，仅供参考