今日论文合集：cs.SD语音15篇，eess.AS音频处理23篇。cs.SD语音【1】 UNSSOR: Unsupervised Neural Speech Separation by Leveraging Over-determined Training Mixtures标题：UNSSOR：利用超定训练混合的无监督神经语音分离链接：https://arxiv.org/abs/2305.20054作者：Zhong-Qiu Wang,Shinji Watanabe机构：Language Technologies Institute, Carnegie Mellon University, Pittsburgh, USA备注：in submission【2】 Multi-Dataset Co-Training with Sharpness-Aware Optimization for Audio Anti-spoofing标题：音频反欺骗的多数据集联合训练与清晰度优化链接：https://arxiv.org/abs/2305.19953作者：Hye-jin Shim,Jee-weon Jung,Tomi Kinnunen机构：University of Eastern Finland, Finland, Carnegie Mellon University, USA备注：Interspeech 2023【3】 Attention-Based Methods For Audio Question Answering标题：基于注意力的音频问答方法链接：https://arxiv.org/abs/2305.19769作者：Parthasaarathy Sudarsanam,Tuomas Virtanen机构：Audio Research Group, Tampere University, Tampere, Finland【4】 Text-to-Speech Pipeline for Swiss German -- A comparison标题：瑞士德语文语转换管道--比较链接：https://arxiv.org/abs/2305.19750作者：Tobias Bollinger,Jan Deriu,Manfred Vogel机构：University of Applied Sciences and Arts Northwestern Switzerland, Windisch, Zurich University of Applied Sciences, Winterthur【5】 XPhoneBERT: A Pre-trained Multilingual Model for Phoneme Representations for Text-to-Speech标题：XPhoneBERT：一种用于文本到语音转换的音素表示的预训练多语言模型链接：https://arxiv.org/abs/2305.19709作者：Linh The Nguyen,Thinh Pham,Dat Quoc Nguyen机构：VinAI Research, Vietnam备注：In Proceedings of INTERSPEECH 2023 (to appear)【6】 Underwater-Art: Expanding Information Perspectives With Text Templates For Underwater Acoustic Target Recognition标题：水下艺术：利用文本模板扩展水声目标识别的信息视角链接：https://arxiv.org/abs/2305.19612作者：Yuan Xie,Jiawei Ren,Ji Xu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【7】 Intelligible Lip-to-Speech Synthesis with Speech Units标题：具有语音单元的可理解的唇语合成链接：https://arxiv.org/abs/2305.19603作者：Jeongsoo Choi,Minsu Kim,Yong Man Ro机构：School of Electrical Engineering, KAIST, South Korea备注：Interspeech 2023【8】 Learning Music Sequence Representation from Text Supervision标题：从文本监督中学习音乐序列表示链接：https://arxiv.org/abs/2305.19602作者：Tianyu Chen,Yuan Xie,Shuai Zhang,Shaohan Huang,Haoyi Zhou,Jianxin Li机构：BDBC, Beihang University, China ,Microsoft Research Asia, China, SKLSDE, Beihang University ,The Institute of Acoustics of the Chinese Academy of Sciences, China备注：None【9】 SVVAD: Personal Voice Activity Detection for Speaker Verification标题：SVVAD：用于说话人确认的个人语音活动检测链接：https://arxiv.org/abs/2305.19581作者：Zuheng Kang,Jianzong Wang,Junqing Peng,Jing Xiao机构：Ping An Technology (Shenzhen) Co., Ltd., Shenzhen, China备注：Accepted by INTERSPEECH 2023【10】 DC CoMix TTS: An End-to-End Expressive TTS with Discrete Code Collaborated with Mixer标题：DC Comix TTS：一种与Mixer合作的端到端可表达的离散代码TTS链接：https://arxiv.org/abs/2305.19567作者：Yerin Choi,Myoung-Wan Koo机构：Sogang University, Korea备注：Accepted at Interspeech 2023【11】 Zero-Shot Automatic Pronunciation Assessment标题：零发音自动测评链接：https://arxiv.org/abs/2305.19563作者：Hongfu Liu,Mingqian Shi,Ye Wang机构：School of Computing, National University of Singapore, Singapore备注：Accepted to Interspeech 2023【12】 Exploring Phonetic Context in Lip Movement for Authentic Talking Face Generation标题：探索唇动中的语音语境以生成真实的说话面孔链接：https://arxiv.org/abs/2305.19556作者：Se Jin Park,Minsu Kim,Jeongsoo Choi,Yong Man Ro机构：Korea Advance Institute of Science & Technology, Daejeon, South Korea【13】 PromptStyle: Controllable Style Transfer for Text-to-Speech with Natural Language Descriptions标题：PromptStyle：自然语言描述文语转换的可控风格转换链接：https://arxiv.org/abs/2305.19522作者：Guanghou Liu,Yongmao Zhang,Yi Lei,Yunlin Chen,Rui Wang,Zhifei Li,Lei Xie机构：School of Computer Science, Northwestern Polytechnical University, Xi’an, China, Shanghai Mobvoi Information Technology Co., Ltd【14】 A Unified Audio-Visual Learning Framework for Localization, Separation, and Recognition标题：用于本地化、分离和识别的统一视听学习框架链接：https://arxiv.org/abs/2305.19458作者：Shentong Mo,Pedro Morgado机构：1Carnegie Mellon University 2University of Wisconsin-Madison, Department of Electrical and Computer Eng【15】 Audio classification using ML methods标题：基于最大似然方法的音频分类链接：https://arxiv.org/abs/2305.19304作者：Krishna Kumar机构：M.Tech Artificial Intelligence, REVA Academy for Corporate, Excellence - RACE, REVA University, Bengaluru, India备注：3 pages, 8 figureseess.AS音频处理【1】 ViLaS: Integrating Vision and Language into Automatic Speech Recognition标题：Vilas：将视觉和语言集成到自动语音识别中链接：https://arxiv.org/abs/2305.19972作者：Minglun Han,Feilong Chen,Ziyi Ni,Linghui Meng,Jing Shi,Shuang Xu,Bo Xu机构：Institute of Automation, Chinese Academy of Sciences, School of Artificial Intelligence, University of Chinese Academy of Sciences, School of Future Technology, University of Chinese Academy of Sciences备注：Speech Recognition with visual-context inputs, Multimodal Speech Recognition, 5 pages【2】 FN-SSL: Full-Band and Narrow-Band Fusion for Sound Source Localization标题：FN-SSL：用于声源定位的全带和窄带融合链接：https://arxiv.org/abs/2305.19610作者：Yabo Wang,Bing Yang,Xiaofei Li机构：Zhejiang University, Hangzhou, China, Westlake Institute for Advanced Study & Westlake University , Hangzhou, China【3】 Few-Shot Speaker Identification Using Lightweight Prototypical Network with Feature Grouping and Interaction标题：基于特征分组和交互的轻量级原型网络的Few-Shot说话人识别链接：https://arxiv.org/abs/2305.19541作者：Yanxiong Li,Hao Chen,Wenchang Cao,Qisheng Huang,Qianhua He机构：few, a task of FSSI is proposed [,], [,]. The FSSI is a newly, emerging task to identify speakers in unlabeled speech samples, (query set) based on few labeled speech samples (support set)., In this paper, we propose a FSSI method using a lightweight备注：12 pages, 4 figures, 12 tables. Accepted for publication in IEEE TMM【4】 Few-shot Class-incremental Audio Classification Using Dynamically Expanded Classifier with Self-attention Modified Prototypes标题：基于自关注修正原型的动态扩展分类器的Few-Shot类增量音频分类链接：https://arxiv.org/abs/2305.19539作者：Yanxiong Li,Wenchang Cao,Wei Xie,Jialong Li,Emmanouil Benetos机构：Electronic and Information Engineering, South China University of Technology, Computer Science, Queen Mary University of London, E,NS London, UK, A. Related Works, Many efforts were made on audio classification with the focus备注：13 pages, 8 figures, 12 tables. Accepted for publication in IEEE TMM【5】 MERLIon CCS Challenge Evaluation Plan标题：鱼尾狮CCS挑战赛评估计划链接：https://arxiv.org/abs/2305.19493作者：Leibny Paola Garcia Perera,Y. H. Victoria Chua,Hexin Liu,Fei Ting Woon,Andy W. H. Khong,Justin Dauwels,Sanjeev Khudanpur,Suzy J. Styles机构：John Hopkins University,Nanyang Technological University,TU Delft, Changelog, Version ,., (,-,-,), • Added amendments to audio segment id naming under Scoring tool and format of, results file and Appendix C.备注：Evaluation plan for Interspeech 2023 special session "MERLIon"【6】 Resource-Efficient Fine-Tuning Strategies for Automatic MOS Prediction in Text-to-Speech for Low-Resource Languages标题：低资源语言文语转换中自动MOS预测的资源高效微调策略链接：https://arxiv.org/abs/2305.19396作者：Phat Do,Matt Coler,Jelske Dijkstra,Esther Klabbers机构：Campus Fryslˆan, University of Groningen, the Netherlands, Fryske AkademyMercator Research Centre, the Netherlands, ReadSpeaker, the Netherlands备注：Accepted at INTERSPEECH 2023【7】 UNSSOR: Unsupervised Neural Speech Separation by Leveraging Over-determined Training Mixtures标题：UNSSOR：利用超定训练混合的无监督神经语音分离链接：https://arxiv.org/abs/2305.20054作者：Zhong-Qiu Wang,Shinji Watanabe机构：Language Technologies Institute, Carnegie Mellon University, Pittsburgh, USA备注：in submission【8】 Multi-Dataset Co-Training with Sharpness-Aware Optimization for Audio Anti-spoofing标题：音频反欺骗的多数据集联合训练与清晰度优化链接：https://arxiv.org/abs/2305.19953作者：Hye-jin Shim,Jee-weon Jung,Tomi Kinnunen机构：University of Eastern Finland, Finland, Carnegie Mellon University, USA备注：Interspeech 2023【9】 Attention-Based Methods For Audio Question Answering标题：基于注意力的音频问答方法链接：https://arxiv.org/abs/2305.19769作者：Parthasaarathy Sudarsanam,Tuomas Virtanen机构：Audio Research Group, Tampere University, Tampere, Finland【10】 Simple yet Effective Code-Switching Language Identification with Multitask Pre-Training and Transfer Learning标题：基于多任务预训练和迁移学习的简单有效语码转换语言识别链接：https://arxiv.org/abs/2305.19759作者：Shuyue Stella Li,Cihan Xiao,Tianjian Li,Bismarck Odoom机构：Center for Language and Speech Processing, Johns Hopkins University备注：8 pages, 3 figures, 7 tables【11】 Text-to-Speech Pipeline for Swiss German -- A comparison标题：瑞士德语文语转换管道--比较链接：https://arxiv.org/abs/2305.19750作者：Tobias Bollinger,Jan Deriu,Manfred Vogel机构：University of Applied Sciences and Arts Northwestern Switzerland, Windisch, Zurich University of Applied Sciences, Winterthur【12】 XPhoneBERT: A Pre-trained Multilingual Model for Phoneme Representations for Text-to-Speech标题：XPhoneBERT：一种用于文本到语音转换的音素表示的预训练多语言模型链接：https://arxiv.org/abs/2305.19709作者：Linh The Nguyen,Thinh Pham,Dat Quoc Nguyen机构：VinAI Research, Vietnam备注：In Proceedings of INTERSPEECH 2023 (to appear)【13】 Underwater-Art: Expanding Information Perspectives With Text Templates For Underwater Acoustic Target Recognition标题：水下艺术：利用文本模板扩展水声目标识别的信息视角链接：https://arxiv.org/abs/2305.19612作者：Yuan Xie,Jiawei Ren,Ji Xu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【14】 Intelligible Lip-to-Speech Synthesis with Speech Units标题：具有语音单元的可理解的唇语合成链接：https://arxiv.org/abs/2305.19603作者：Jeongsoo Choi,Minsu Kim,Yong Man Ro机构：School of Electrical Engineering, KAIST, South Korea备注：Interspeech 2023【15】 Learning Music Sequence Representation from Text Supervision标题：从文本监督中学习音乐序列表示链接：https://arxiv.org/abs/2305.19602作者：Tianyu Chen,Yuan Xie,Shuai Zhang,Shaohan Huang,Haoyi Zhou,Jianxin Li机构：BDBC, Beihang University, China ,Microsoft Research Asia, China, SKLSDE, Beihang University ,The Institute of Acoustics of the Chinese Academy of Sciences, China备注：None【16】 The Tag-Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR标题：标签团队方法：利用CLS和语言标签来增强多语言ASR链接：https://arxiv.org/abs/2305.19584作者：Kaousheik Jayakumar,Vrunda N. Sukhadia,A Arunkumar,S. Umesh机构：Speech Lab, Indian Institute of Technology Madras, India备注：5 pages,5 figures, submitted to INTERSPEECH2023【17】 SVVAD: Personal Voice Activity Detection for Speaker Verification标题：SVVAD：用于说话人确认的个人语音活动检测链接：https://arxiv.org/abs/2305.19581作者：Zuheng Kang,Jianzong Wang,Junqing Peng,Jing Xiao机构：Ping An Technology (Shenzhen) Co., Ltd., Shenzhen, China备注：Accepted by INTERSPEECH 2023【18】 DC CoMix TTS: An End-to-End Expressive TTS with Discrete Code Collaborated with Mixer标题：DC Comix TTS：一种与Mixer合作的端到端可表达的离散代码TTS链接：https://arxiv.org/abs/2305.19567作者：Yerin Choi,Myoung-Wan Koo机构：Sogang University, Korea备注：Accepted at Interspeech 2023【19】 Zero-Shot Automatic Pronunciation Assessment标题：零发音自动测评链接：https://arxiv.org/abs/2305.19563作者：Hongfu Liu,Mingqian Shi,Ye Wang机构：School of Computing, National University of Singapore, Singapore备注：Accepted to Interspeech 2023【20】 Exploring Phonetic Context in Lip Movement for Authentic Talking Face Generation标题：探索唇动中的语音语境以生成真实的说话面孔链接：https://arxiv.org/abs/2305.19556作者：Se Jin Park,Minsu Kim,Jeongsoo Choi,Yong Man Ro机构：Korea Advance Institute of Science & Technology, Daejeon, South Korea【21】 PromptStyle: Controllable Style Transfer for Text-to-Speech with Natural Language Descriptions标题：PromptStyle：自然语言描述文语转换的可控风格转换链接：https://arxiv.org/abs/2305.19522作者：Guanghou Liu,Yongmao Zhang,Yi Lei,Yunlin Chen,Rui Wang,Zhifei Li,Lei Xie机构：School of Computer Science, Northwestern Polytechnical University, Xi’an, China, Shanghai Mobvoi Information Technology Co., Ltd【22】 A Unified Audio-Visual Learning Framework for Localization, Separation, and Recognition标题：用于本地化、分离和识别的统一视听学习框架链接：https://arxiv.org/abs/2305.19458作者：Shentong Mo,Pedro Morgado机构：1Carnegie Mellon University 2University of Wisconsin-Madison, Department of Electrical and Computer Eng【23】 Audio classification using ML methods标题：基于最大似然方法的音频分类链接：https://arxiv.org/abs/2305.19304作者：Krishna Kumar机构：M.Tech Artificial Intelligence, REVA Academy for Corporate, Excellence - RACE, REVA University, Bengaluru, India备注：3 pages, 8 figures机器翻译，仅供参考
每日语音速递
