今日论文合集：cs.SD语音16篇，eess.AS音频处理17篇。cs.SD语音【1】 Video-to-Music Recommendation using Temporal Alignment of Segments标题：使用片段的时间对齐的视频到音乐推荐链接：https://arxiv.org/abs/2306.07187作者：Laure Prétet,Gaël Richard,Clément Souchier,Geoffroy Peeters备注：None【2】 Exploring Attention Mechanisms for Multimodal Emotion Recognition in an Emergency Call Center Corpus标题：在紧急呼叫中心语料库中探索多通道情绪识别的注意机制链接：https://arxiv.org/abs/2306.07115作者：Théo Deschamps-Berger,Lori Lamel,Laurence Devillers机构：⋆ LISN - CNRS, Paris-Saclay University, Orsay, France, † LISN - CNRS, Sorbonne University, Orsay, France备注：None【3】 Underwater Acoustic Target Recognition based on Smoothness-inducing Regularization and Spectrogram-based Data Augmentation标题：基于平滑诱导正则化和谱图数据增强的水声目标识别链接：https://arxiv.org/abs/2306.06945作者：Ji Xu,Yuan Xie,Wenchao Wang机构：Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics, Chinese Academy of Sciences, University of Chinese Academy of Sciences备注：None【4】 Estimating the Uncertainty in Emotion Attributes using Deep Evidential Regression标题：基于深度证据回归的情感属性不确定性估计链接：https://arxiv.org/abs/2306.06760作者：Wen Wu,Chao Zhang,Philip C. Woodland机构：Department of Engineering, University of Cambridge, Cambridge, UK, Department of Electrical Engineering, Tsinghua University, Beijing, China备注：Accepted by ACL 2023【5】 Mandarin Electrolaryngeal Speech Voice Conversion using Cross-domain Features标题：基于跨域特征的普通话电喉语音转换链接：https://arxiv.org/abs/2306.06653作者：Hsin-Hao Chen,Yung-Lun Chien,Ming-Chi Yen,Shu-Wei Tsai,Yu Tsao,Tai-shih Chi,Hsin-Min Wang机构：National Yang Ming Chiao Tung University, Academia Sinica, National Cheng Kung University Hospital备注：Accepted to INTERSPEECH 2023【6】 Audio-Visual Mandarin Electrolaryngeal Speech Voice Conversion标题：视听普通话电喉语音转换链接：https://arxiv.org/abs/2306.06652作者：Yung-Lun Chien,Hsin-Hao Chen,Ming-Chi Yen,Shu-Wei Tsai,Hsin-Min Wang,Yu Tsao,Tai-Shih Chi机构：National Yang Ming Chiao Tung University, Academia Sinica, National Cheng Kung University Hospital备注：Accepted to INTERSPEECH 2023【7】 High-Fidelity Audio Compression with Improved RVQGAN标题：采用改进的RVQGAN的高保真音频压缩链接：https://arxiv.org/abs/2306.06546作者：Rithesh Kumar,Prem Seetharaman,Alejandro Luebs,Ishaan Kumar,Kundan Kumar机构：Descript, Inc.【8】 Vocoder-Free Non-Parallel Conversion of Whispered Speech With Masked Cycle-Consistent Generative Adversarial Networks标题：基于掩蔽循环一致生成对抗性网络的无声码非并行耳语转换链接：https://arxiv.org/abs/2306.06514作者：Dominik Wagner,Ilja Baumann,Tobias Bocklet机构：Technische Hochschule N¨urnberg Georg Simon Ohm, Germany, Intel Labs【9】 Everybody Compose: Deep Beats To Music标题：大家作曲：低沉的音乐节拍链接：https://arxiv.org/abs/2306.06284作者：Conghao Shen,Violet Z. Yao,Yixin Liu机构：Stanford University, United States备注：None【10】 Probing self-supervised speech models for phonetic and phonemic information: a case study in aspiration标题：探索语音和音素信息的自我监督语音模型：以《吸气》为例链接：https://arxiv.org/abs/2306.06232作者：Kinan Martin,Jon Gauthier,Canaan Breiss,Roger Levy机构：Massachusetts Institute of Technology, USA备注：Accepted at INTERSPEECH 2023【11】 Parameter-efficient Dysarthric Speech Recognition Using Adapter Fusion and Householder Transformation标题：基于适配器融合和Household变换的参数高效节律语音识别链接：https://arxiv.org/abs/2306.07090作者：Jinzi Qi,Hugo Van hamme机构：Department Electrical Engineering-ESAT-PSI, KULeuven, Belgium备注：Accepted by Interspeech 2023【12】 Multi-View Frequency-Attention Alternative to CNN Frontends for Automatic Speech Recognition标题：用于自动语音识别的CNN前沿的多视点频率注意替代链接：https://arxiv.org/abs/2306.06954作者：Belen Alastruey,Lukas Drude,Jahn Heymann,Simon Wiesler机构：TALP Research Center, Universitat Politecnica de Catalunya, Barcelona, Spain, Amazon Alexa, Germany【13】 HiddenSinger: High-Quality Singing Voice Synthesis via Neural Audio Codec and Latent Diffusion Models标题：HiddenSinger：基于神经音频编解码器和潜在扩散模型的高质量歌声合成链接：https://arxiv.org/abs/2306.06814作者：Ji-Sang Hwang,Sang-Hoon Lee,Seong-Whan Lee机构：Fellow, IEEE备注：11 pages, 5 figures, 5 tables, under review【14】 What Can an Accent Identifier Learn? Probing Phonetic and Prosodic Information in a Wav2vec2-based Accent Identification Model标题：口音识别器能学到什么？基于Wav2ve2的口音识别模型中语音和韵律信息的探讨链接：https://arxiv.org/abs/2306.06524作者：Mu Yang,Ram C. M. C. Shekar,Okim Kang,John H. L. Hansen机构：Center for Robust Speech Systems (CRSS), University of Texas at Dallas, Richardson, TX, USA, Northern Arizona University, Flagstaff, AZ, USA备注：Accepted by Interspeech 2023【15】 Audio-Visual Speech Enhancement With Selective Off-Screen Speech Extraction标题：基于选择性屏外语音提取的视听语音增强链接：https://arxiv.org/abs/2306.06495作者：Tomoya Yoshinaga,Keitaro Tanaka,Shigeo Morishima机构：School of Advanced Science and Engineering, Waseda University, Tokyo, Japan, Waseda Research Institute for Science and Engineering, Tokyo, Japan备注：Accepted by EUSIPCO 2023【16】 Semi-supervsied Learning-based Sound Event Detection using Freuqency Dynamic Convolution with Large Kernel Attention for DCASE Challenge 2023 Task 4标题：DCASE挑战赛2023任务4基于半监督学习的大核注意力动态卷积声音事件检测链接：https://arxiv.org/abs/2306.06461作者：Ji Won Kim,Sang Won Son,Yoonah Song,Hong Kook Kim,Il Hoon Song,Jeong Eun Lim机构：AI Graduate School, School of EECS, Gwangju Institute of Science and Technology, Gwangju , Korea, AI Lab., R&D Center, Hanwha Vision, Seongnam-si, Gyeonggi-do , Korea备注：DCASE 2023 Challenge Task 4A, 5 pageseess.AS音频处理【1】 Parameter-efficient Dysarthric Speech Recognition Using Adapter Fusion and Householder Transformation标题：基于适配器融合和Household变换的参数高效节律语音识别链接：https://arxiv.org/abs/2306.07090*与cs.SD语音【11】为同一篇作者：Jinzi Qi,Hugo Van hamme机构：Department Electrical Engineering-ESAT-PSI, KULeuven, Belgium备注：Accepted by Interspeech 2023【2】 Multi-View Frequency-Attention Alternative to CNN Frontends for Automatic Speech Recognition标题：用于自动语音识别的CNN前沿的多视点频率注意替代链接：https://arxiv.org/abs/2306.06954*与cs.SD语音【12】为同一篇作者：Belen Alastruey,Lukas Drude,Jahn Heymann,Simon Wiesler机构：TALP Research Center, Universitat Politecnica de Catalunya, Barcelona, Spain, Amazon Alexa, Germany【3】 HiddenSinger: High-Quality Singing Voice Synthesis via Neural Audio Codec and Latent Diffusion Models标题：HiddenSinger：基于神经音频编解码器和潜在扩散模型的高质量歌声合成链接：https://arxiv.org/abs/2306.06814*与cs.SD语音【13】为同一篇作者：Ji-Sang Hwang,Sang-Hoon Lee,Seong-Whan Lee机构：Fellow, IEEE备注：11 pages, 5 figures, 5 tables, under review【4】 What Can an Accent Identifier Learn? Probing Phonetic and Prosodic Information in a Wav2vec2-based Accent Identification Model标题：口音识别器能学到什么？基于Wav2ve2的口音识别模型中语音和韵律信息的探讨链接：https://arxiv.org/abs/2306.06524*与cs.SD语音【14】为同一篇作者：Mu Yang,Ram C. M. C. Shekar,Okim Kang,John H. L. Hansen机构：Center for Robust Speech Systems (CRSS), University of Texas at Dallas, Richardson, TX, USA, Northern Arizona University, Flagstaff, AZ, USA备注：Accepted by Interspeech 2023【5】 Audio-Visual Speech Enhancement With Selective Off-Screen Speech Extraction标题：基于选择性屏外语音提取的视听语音增强链接：https://arxiv.org/abs/2306.06495*与cs.SD语音【15】为同一篇作者：Tomoya Yoshinaga,Keitaro Tanaka,Shigeo Morishima机构：School of Advanced Science and Engineering, Waseda University, Tokyo, Japan, Waseda Research Institute for Science and Engineering, Tokyo, Japan备注：Accepted by EUSIPCO 2023【6】 Semi-supervsied Learning-based Sound Event Detection using Freuqency Dynamic Convolution with Large Kernel Attention for DCASE Challenge 2023 Task 4标题：DCASE挑战赛2023任务4基于半监督学习的大核注意力动态卷积声音事件检测链接：https://arxiv.org/abs/2306.06461*与cs.SD语音【16】为同一篇作者：Ji Won Kim,Sang Won Son,Yoonah Song,Hong Kook Kim,Il Hoon Song,Jeong Eun Lim机构：AI Graduate School, School of EECS, Gwangju Institute of Science and Technology, Gwangju , Korea, AI Lab., R&D Center, Hanwha Vision, Seongnam-si, Gyeonggi-do , Korea备注：DCASE 2023 Challenge Task 4A, 5 pages【7】 Video-to-Music Recommendation using Temporal Alignment of Segments标题：使用片段的时间对齐的视频到音乐推荐链接：https://arxiv.org/abs/2306.07187*与cs.SD语音【1】为同一篇作者：Laure Prétet,Gaël Richard,Clément Souchier,Geoffroy Peeters备注：None【8】 Exploring Attention Mechanisms for Multimodal Emotion Recognition in an Emergency Call Center Corpus标题：在紧急呼叫中心语料库中探索多通道情绪识别的注意机制链接：https://arxiv.org/abs/2306.07115*与cs.SD语音【2】为同一篇作者：Théo Deschamps-Berger,Lori Lamel,Laurence Devillers机构：⋆ LISN - CNRS, Paris-Saclay University, Orsay, France, † LISN - CNRS, Sorbonne University, Orsay, France备注：None【9】 Multimodal Audio-textual Architecture for Robust Spoken Language Understanding标题：面向健壮口语理解的多通道音频-文本结构链接：https://arxiv.org/abs/2306.06819作者：Anderson R. Avila,Mehdi Rezagholizadeh,Chao Xing机构：INRS-EMT, Montréal, Québec, Canada, Noah’s Ark Lab【10】 Estimating the Uncertainty in Emotion Attributes using Deep Evidential Regression标题：基于深度证据回归的情感属性不确定性估计链接：https://arxiv.org/abs/2306.06760*与cs.SD语音【4】为同一篇作者：Wen Wu,Chao Zhang,Philip C. Woodland机构：Department of Engineering, University of Cambridge, Cambridge, UK, Department of Electrical Engineering, Tsinghua University, Beijing, China备注：Accepted by ACL 2023【11】 Reducing Barriers to Self-Supervised Learning: HuBERT Pre-training with Academic Compute标题：减少自我监督学习的障碍：休BERT用学术计算进行预训链接：https://arxiv.org/abs/2306.06672作者：William Chen,Xuankai Chang,Yifan Peng,Zhaoheng Ni,Soumi Maiti,Shinji Watanabe机构：Language Technologies Institute, Carnegie Mellon University, USA, Electrical and Computer Engineering, Carnegie Mellon University, USA, Meta, USA备注：Accepted at INTERSPEECH 2023【12】 Mandarin Electrolaryngeal Speech Voice Conversion using Cross-domain Features标题：基于跨域特征的普通话电喉语音转换链接：https://arxiv.org/abs/2306.06653*与cs.SD语音【5】为同一篇作者：Hsin-Hao Chen,Yung-Lun Chien,Ming-Chi Yen,Shu-Wei Tsai,Yu Tsao,Tai-shih Chi,Hsin-Min Wang机构：National Yang Ming Chiao Tung University, Academia Sinica, National Cheng Kung University Hospital备注：Accepted to INTERSPEECH 2023【13】 Audio-Visual Mandarin Electrolaryngeal Speech Voice Conversion标题：视听普通话电喉语音转换链接：https://arxiv.org/abs/2306.06652*与cs.SD语音【6】为同一篇作者：Yung-Lun Chien,Hsin-Hao Chen,Ming-Chi Yen,Shu-Wei Tsai,Hsin-Min Wang,Yu Tsao,Tai-Shih Chi机构：National Yang Ming Chiao Tung University, Academia Sinica, National Cheng Kung University Hospital备注：Accepted to INTERSPEECH 2023【14】 High-Fidelity Audio Compression with Improved RVQGAN标题：采用改进的RVQGAN的高保真音频压缩链接：https://arxiv.org/abs/2306.06546*与cs.SD语音【7】为同一篇作者：Rithesh Kumar,Prem Seetharaman,Alejandro Luebs,Ishaan Kumar,Kundan Kumar机构：Descript, Inc.【15】 Vocoder-Free Non-Parallel Conversion of Whispered Speech With Masked Cycle-Consistent Generative Adversarial Networks标题：基于掩蔽循环一致生成对抗性网络的无声码非并行耳语转换链接：https://arxiv.org/abs/2306.06514*与cs.SD语音【8】为同一篇作者：Dominik Wagner,Ilja Baumann,Tobias Bocklet机构：Technische Hochschule N¨urnberg Georg Simon Ohm, Germany, Intel Labs【16】 Everybody Compose: Deep Beats To Music标题：大家作曲：低沉的音乐节拍链接：https://arxiv.org/abs/2306.06284*与cs.SD语音【9】为同一篇作者：Conghao Shen,Violet Z. Yao,Yixin Liu机构：Stanford University, United States备注：None【17】 Probing self-supervised speech models for phonetic and phonemic information: a case study in aspiration标题：探索语音和音素信息的自我监督语音模型：以《吸气》为例链接：https://arxiv.org/abs/2306.06232*与cs.SD语音【10】为同一篇作者：Kinan Martin,Jon Gauthier,Canaan Breiss,Roger Levy机构：Massachusetts Institute of Technology, USA备注：Accepted at INTERSPEECH 2023机器翻译，仅供参考
每日语音速递
