今日论文合集：cs.SD语音14篇，eess.AS音频处理18篇。cs.SD语音【1】 Self-supervised learning with diffusion-based multichannel speech enhancement for speaker verification under noisy conditions标题：基于扩散的多通道语音增强自监督学习在噪声环境下的说话人确认链接：https://arxiv.org/abs/2307.02244作者：Sandipana Dowerah,Ajinkya Kulkarni,Romain Serizel,Denis Jouvet机构：∗Universit´e de Lorraine, CNRS, Inria, Loria, F-, Nancy, France, †MBZUAI University, UAE【2】 LOAF-M2L: Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation标题：Loaf-M2L：可唱旋律到歌词生成的措辞和格式的联合学习链接：https://arxiv.org/abs/2307.02146作者：Longshen Ou,Xichu Ma,Ye Wang机构：National University of singapore备注：An extension of our previous work arXiv:2305.16816 [cs.CL]【3】 Going Retro: Astonishingly Simple Yet Effective Rule-based Prosody Modelling for Speech Synthesis Simulating Emotion Dimensions标题：回溯：模拟情感维度的语音合成的基于规则的韵律建模链接：https://arxiv.org/abs/2307.02132作者：Felix Burkhardt,Uwe Reichel,Florian Eyben,Björn Schuller机构：audEERING GmbH, Germany, Chair EIHW, University of Augsburg, Germany, GLAM, Imperial College London, UK备注：accepted at 34th ESSV 2023, Munich 2023【4】 Align With Purpose: Optimize Desired Properties in CTC Models with a General Plug-and-Play Framework标题：与目的一致：使用通用即插即用框架优化CTC模型中所需的属性链接：https://arxiv.org/abs/2307.01715作者：Eliya Segev,Maya Alroy,Ronen Katsir,Noam Wies,Ayana Shenhav,Yael Ben-Oren,David Zar,Oren Tadmor,Jacob Bitterman,Amnon Shashua,Tal Rosenwein机构：OrCam Technologies LTD, Jerusalem, Israel【5】 Pretraining Conformer with ASR or ASV for Anti-Spoofing Countermeasure标题：使用ASR或ASV进行反欺骗对策的预训练一致性链接：https://arxiv.org/abs/2307.01546作者：Yikang Wang,Hiromitsu Nishizaki,Ming Li机构：Integrated Graduate School of Medicine, Engineering, and Agricultural Sciences, University of Yamanashi, Japan, Data Science Research Center, Duke Kunshan University, China, School of Computer Science, Wuhan University, China备注：6 pages, 3 figures【6】 Spatial-temporal Graph Based Multi-channel Speaker Verification With Ad-hoc Microphone Arrays标题：基于时空图的自组织麦克风阵列多声道说话人确认链接：https://arxiv.org/abs/2307.01386作者：Yijiang Chen,Chengdong Liang,Xiao-Lei Zhang【7】 Semantic enrichment towards efficient speech representations标题：面向高效语音表示的语义丰富链接：https://arxiv.org/abs/2307.01323作者：Gaëlle Laperrière,Ha Nguyen,Sahar Ghannay,Bassam Jabaian,Yannick Estève机构：Esteve 1 1LIA - Avignon Universit´e, France 2LISN - CNRSUniversit´e Paris-Saclay备注：INTERSPEECH 2023【8】 RobustL2S: Speaker-Specific Lip-to-Speech Synthesis exploiting Self-Supervised Representations标题：RobustL2S：基于自监督表示的特定于说话人的唇语转语音合成链接：https://arxiv.org/abs/2307.01233作者：Neha Sahipjohn,Neil Shah,Vishal Tambrahalli,Vineet Gandhi机构：∗ CVIT, Kohli Centre for Intelligent Systems, IIIT Hyderabad, India, † TCS Research, Pune, India【9】 EmoGen: Eliminating Subjective Bias in Emotional Music Generation标题：EmoGen：消除情感音乐生成中的主观偏见链接：https://arxiv.org/abs/2307.01229作者：Chenfei Kang,Peiling Lu,Botao Yu,Xu Tan,Wei Ye,Shikun Zhang,Jiang Bian机构：Equalcontribution 1ShanghaiJiaoTongUniversity, China 2Microsoft Research Asia 3Nanjing University, Peking University备注：12 pages, 7 pages【10】 Of Spiky SVDs and Music Recommendation标题：尖峰SVD和音乐推荐链接：https://arxiv.org/abs/2307.01212作者：Darius Afchar,Romain Hennequin,Vincent Guigue机构：MLIA – Sorbonne Universités备注：Accepted for RecSys 2023 (Singapour, 18-22 September)【11】 Why can big.bi be changed to bi.gbi? A mathematical model of syllabification and articulatory synthesis标题：为什么Big.bi可以改为bi.gbi？音节和发音合成的数学模型链接：https://arxiv.org/abs/2307.02299作者：Frédéric Berthommier备注：5 pages, 4 figures, submitted to INTERSPEECH 2023【12】 A Database with Directivities of Musical Instruments标题：一个具有指导性的乐器数据库链接：https://arxiv.org/abs/2307.02110作者：David Ackermann,Fabian Brinkmann,Stefan Weinzierl机构：AND, Audio Communication Group, Technische Universit¨at Berlin, Germany【13】 Flowchase: a Mobile Application for Pronunciation Training标题：Flowchase：一款语音训练手机应用链接：https://arxiv.org/abs/2307.02051作者：Noé Tits,Zoé Broisson备注：Interspeech 2023 - SLaTE workshop 2023 (Speech and Language Technology in Education)【14】 Disentanglement in a GAN for Unconditional Speech Synthesis标题：用于无条件语音合成的GaN中的解缠链接：https://arxiv.org/abs/2307.01673作者：Matthew Baas,Herman Kamper机构：Matthew Baas and Herman Kamper are with the Department of Electricaland Electronic Engineering, Stellenbosch University备注：12 pages, 5 tables, 4 figures. Submitted to IEEE TASLP. arXiv admin note: substantial text overlap with arXiv:2210.05271eess.AS音频处理【1】 Deep Speech Synthesis from MRI-Based Articulatory Representations标题：基于MRI的节律表示的深度语音合成链接：https://arxiv.org/abs/2307.02471作者：Peter Wu,Tingle Li,Yijing Lu,Yubin Zhang,Jiachen Lian,Alan W Black,Louis Goldstein,Shinji Watanabe,Gopala K. Anumanchipalli机构：University of California, Berkeley, United States, Carnegie Mellon University, United States, University of Southern California, United States【2】 Online Hybrid CTC/Attention End-to-End Automatic Speech Recognition Architecture标题：在线混合CTC/注意端到端自动语音识别架构链接：https://arxiv.org/abs/2307.02351作者：Haoran Miao,Gaofeng Cheng,Pengyuan Zhang,Yonghong Yan机构：cn) are with Institute of Acous-tics, Chinese Academy of Sciences and University of Chinese Academy ofSciences备注：None【3】 Why can big.bi be changed to bi.gbi? A mathematical model of syllabification and articulatory synthesis标题：为什么Big.bi可以改为bi.gbi？音节和发音合成的数学模型链接：https://arxiv.org/abs/2307.02299作者：Frédéric Berthommier备注：5 pages, 4 figures, submitted to INTERSPEECH 2023【4】 Differentially Private Adversarial Auto-Encoder to Protect Gender in Voice Biometrics标题：语音生物识别中保护性别的差异化对抗性自动编码器链接：https://arxiv.org/abs/2307.02135作者：Oubaïda Chouchane,Michele Panariello,Oualid Zari,Ismet Kerenciler,Imen Chihaoui,Massimiliano Todisco,Melek Önen机构：EURECOM, Sophia Antipolis, France备注：6 Pages, 3 figures【5】 A Database with Directivities of Musical Instruments标题：一个具有指导性的乐器数据库链接：https://arxiv.org/abs/2307.02110作者：David Ackermann,Fabian Brinkmann,Stefan Weinzierl机构：AND, Audio Communication Group, Technische Universit¨at Berlin, Germany【6】 Leveraging multilingual transfer for unsupervised semantic acoustic word embeddings标题：利用多语言迁移实现无监督语义声学单词嵌入链接：https://arxiv.org/abs/2307.02083作者：Christiaan Jacobs,Herman Kamper机构：Stellenbosch University备注：Submitted to IEEE SPL【7】 Flowchase: a Mobile Application for Pronunciation Training标题：Flowchase：一款语音训练手机应用链接：https://arxiv.org/abs/2307.02051作者：Noé Tits,Zoé Broisson备注：Interspeech 2023 - SLaTE workshop 2023 (Speech and Language Technology in Education)【8】 Disentanglement in a GAN for Unconditional Speech Synthesis标题：用于无条件语音合成的GaN中的解缠链接：https://arxiv.org/abs/2307.01673作者：Matthew Baas,Herman Kamper机构：Matthew Baas and Herman Kamper are with the Department of Electricaland Electronic Engineering, Stellenbosch University备注：12 pages, 5 tables, 4 figures. Submitted to IEEE TASLP. arXiv admin note: substantial text overlap with arXiv:2210.05271【9】 Self-supervised learning with diffusion-based multichannel speech enhancement for speaker verification under noisy conditions标题：基于扩散的多通道语音增强自监督学习在噪声环境下的说话人确认链接：https://arxiv.org/abs/2307.02244作者：Sandipana Dowerah,Ajinkya Kulkarni,Romain Serizel,Denis Jouvet机构：∗Universit´e de Lorraine, CNRS, Inria, Loria, F-, Nancy, France, †MBZUAI University, UAE【10】 LOAF-M2L: Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation标题：Loaf-M2L：可唱旋律到歌词生成的措辞和格式的联合学习链接：https://arxiv.org/abs/2307.02146作者：Longshen Ou,Xichu Ma,Ye Wang机构：National University of singapore备注：An extension of our previous work arXiv:2305.16816 [cs.CL]【11】 Going Retro: Astonishingly Simple Yet Effective Rule-based Prosody Modelling for Speech Synthesis Simulating Emotion Dimensions标题：回溯：模拟情感维度的语音合成的基于规则的韵律建模链接：https://arxiv.org/abs/2307.02132作者：Felix Burkhardt,Uwe Reichel,Florian Eyben,Björn Schuller机构：audEERING GmbH, Germany, Chair EIHW, University of Augsburg, Germany, GLAM, Imperial College London, UK备注：accepted at 34th ESSV 2023, Munich 2023【12】 Align With Purpose: Optimize Desired Properties in CTC Models with a General Plug-and-Play Framework标题：与目的一致：使用通用即插即用框架优化CTC模型中所需的属性链接：https://arxiv.org/abs/2307.01715作者：Eliya Segev,Maya Alroy,Ronen Katsir,Noam Wies,Ayana Shenhav,Yael Ben-Oren,David Zar,Oren Tadmor,Jacob Bitterman,Amnon Shashua,Tal Rosenwein机构：OrCam Technologies LTD, Jerusalem, Israel【13】 Pretraining Conformer with ASR or ASV for Anti-Spoofing Countermeasure标题：使用ASR或ASV进行反欺骗对策的预训练一致性链接：https://arxiv.org/abs/2307.01546作者：Yikang Wang,Hiromitsu Nishizaki,Ming Li机构：Integrated Graduate School of Medicine, Engineering, and Agricultural Sciences, University of Yamanashi, Japan, Data Science Research Center, Duke Kunshan University, China, School of Computer Science, Wuhan University, China备注：6 pages, 3 figures【14】 Spatial-temporal Graph Based Multi-channel Speaker Verification With Ad-hoc Microphone Arrays标题：基于时空图的自组织麦克风阵列多声道说话人确认链接：https://arxiv.org/abs/2307.01386作者：Yijiang Chen,Chengdong Liang,Xiao-Lei Zhang【15】 Semantic enrichment towards efficient speech representations标题：面向高效语音表示的语义丰富链接：https://arxiv.org/abs/2307.01323作者：Gaëlle Laperrière,Ha Nguyen,Sahar Ghannay,Bassam Jabaian,Yannick Estève机构：Esteve 1 1LIA - Avignon Universit´e, France 2LISN - CNRSUniversit´e Paris-Saclay备注：INTERSPEECH 2023【16】 RobustL2S: Speaker-Specific Lip-to-Speech Synthesis exploiting Self-Supervised Representations标题：RobustL2S：基于自监督表示的特定于说话人的唇语转语音合成链接：https://arxiv.org/abs/2307.01233作者：Neha Sahipjohn,Neil Shah,Vishal Tambrahalli,Vineet Gandhi机构：∗ CVIT, Kohli Centre for Intelligent Systems, IIIT Hyderabad, India, † TCS Research, Pune, India【17】 EmoGen: Eliminating Subjective Bias in Emotional Music Generation标题：EmoGen：消除情感音乐生成中的主观偏见链接：https://arxiv.org/abs/2307.01229作者：Chenfei Kang,Peiling Lu,Botao Yu,Xu Tan,Wei Ye,Shikun Zhang,Jiang Bian机构：Equalcontribution 1ShanghaiJiaoTongUniversity, China 2Microsoft Research Asia 3Nanjing University, Peking University备注：12 pages, 7 pages【18】 Of Spiky SVDs and Music Recommendation标题：尖峰SVD和音乐推荐链接：https://arxiv.org/abs/2307.01212作者：Darius Afchar,Romain Hennequin,Vincent Guigue机构：MLIA – Sorbonne Universités备注：Accepted for RecSys 2023 (Singapour, 18-22 September)机器翻译，仅供参考
每日语音速递
