语音/音频处理学术速递[5.30]

今日论文合集：cs.SD语音25篇，eess.AS音频处理31篇。

cs.SD语音

【1】 Exploration of Efficient End-to-End ASR using Discretized Input from Self-Supervised Learning标题：利用离散化的自监督学习输入探索高效的端到端ASR链接：https://arxiv.org/abs/2305.18108

作者：Xuankai Chang,Brian Yan,Yuya Fujita,Takashi Maekaku,Shinji Watanabe机构：Carnegie Mellon University, PA, USA, Yahoo Japan Corporation, Japan备注：Accepted at INTERSPEECH 2023

【2】 Few-shot Class-incremental Audio Classification Using Adaptively-refined Prototypes标题：基于自适应细化原型的Few-Shot类增量音频分类链接：https://arxiv.org/abs/2305.18045

作者：Wei Xie,Yanxiong Li,Qianhua He,Wenchang Cao,Tuomas Virtanen机构：School of Electronic and Information Engineering, South China University of Technology, Guangzhou , China, Audio Research Group, Tampere University, Tampere, Finland备注：5 pages,2 figures, Accepted by Interspeech 2023

【3】 ADAPTERMIX: Exploring the Efficacy of Mixture of Adapters for Low-Resource TTS Adaptation标题：ADAPTERMIX：探索混合适配器对低资源TTS适应的有效性链接：https://arxiv.org/abs/2305.18028

作者：Ambuj Mehrish,Abhinav Ramesh Kashyap,Li Yingting,Navonil Majumder,Soujanya Poria机构：Singapore University of Technology and Design, Singapore, ASUS Intelligent Cloud Services (AICS) Singapore, Beijing University of Posts and Telecommunications, China备注：Interspeech 2023

【4】 Can We Trust Explainable AI Methods on ASR? An Evaluation on Phoneme Recognition标题：我们能在ASR上信任可解释的人工智能方法吗？音素识别研究述评链接：https://arxiv.org/abs/2305.18011

作者：Xiaoliang Wu,Peter Bell,Ajitha Rajan机构：School of Informatics, University of Edinburgh

【5】 Multi-Scale Attention for Audio Question Answering标题：音频问答中的多尺度注意链接：https://arxiv.org/abs/2305.17993

作者：Guangyao Li,Yixin Xu,Di Hu机构：Gaoling School of Artificial Intelligence, Renmin Uniiversity of China, Beijing, Beijing Key Laboratory of Big Data Management and Analysis Methods, Beijing备注：Accepted by InterSpeech 2023

【6】 speech and noise dual-stream spectrogram refine network with speech distortion loss for robust speech recognition标题：用于稳健语音识别的语音和噪声双流谱图精化语音失真损失网络链接：https://arxiv.org/abs/2305.17860

作者：Haoyu Lu,Nan Li,Tongtong Song,Longbiao Wang,Jianwu Dang,Xiaobao Wang,Shiliang Zhang机构：Tianjin Key Laboratory of Cognitive Computing and Application, College of Intelligence and Computing, Tianjin University, Tianjin, China

【7】 Retraining-free Customized ASR for Enharmonic Words Based on a Named-Entity-Aware Model and Phoneme Similarity Estimation标题：基于命名实体感知模型和音素相似度估计的免再训练非调和词定制ASR链接：https://arxiv.org/abs/2305.17846

作者：Yui Sudo,Kazuya Hata,Kazuhiro Nakadai机构：Honda Research Institute Japan Co., Ltd., Saitama, Japan, Department of Systems and Control Engineering, School of Engineering, Tokyo Institute of Technology, Tokyo, Japan备注：accepted by INTERSPEECH2023

【8】 Streaming Audio Transformers for Online Audio Tagging标题：用于在线音频标记的流音频转换器链接：https://arxiv.org/abs/2305.17834

作者：Heinrich Dinkel,Zhiyong Yan,Yongqing Wang,Junbo Zhang,Yujun Wang机构：Xiaomi Corporation, Beijing, China

【9】 RASR2: The RWTH ASR Toolkit for Generic Sequence-to-sequence Speech Recognition标题：RASR2：用于通用序列到序列语音识别的RWTH ASR工具包链接：https://arxiv.org/abs/2305.17782

作者：Wei Zhou,Eugen Beck,Simon Berger,Ralf Schlüter,Hermann Ney机构：Machine Learning and Human Language Technology, Computer Science Department, RWTH Aachen University, Aachen, Germany, AppTek GmbH, Aachen, Germany备注：accepted at Interspeech 2023

【10】 CAPTDURE: Captioned Sound Dataset of Single Sources标题：CAPTDURE：单源字幕声音数据集链接：https://arxiv.org/abs/2305.17758

作者：Yuki Okamoto,Kanta Shimonishi,Keisuke Imoto,Kota Dohi,Shota Horiguchi,Yohei Kawaguchi机构：Ritsumeikan University, Japan, Doshisha University, Japan, Hitachi, Ltd., Japan备注：Accepted to INTERSPEECH2023

【11】 Bayesian inference and neural estimation of acoustic wave propagation标题：声波传播的贝叶斯推断和神经估计链接：https://arxiv.org/abs/2305.17749

作者：Yongchao Huang,Yuhang He,Hong Ge机构： Department of Computer Science, University of Oxford‡hg 3 4 4

【12】 Range-Based Equal Error Rate for Spoof Localization标题：基于测距的等错误率欺骗定位方法链接：https://arxiv.org/abs/2305.17739

作者：Lin Zhang,Xin Wang,Erica Cooper,Nicholas Evans,Junichi Yamagishi机构：National Institute of Informatics, Tokyo, Japan, SOKENDAI (The Graduate University for Advanced Studies), Kanagawa, Japan, Digital Security Department, EURECOM, France备注：Accepted to Interspeech 2023

【13】 Investigating Pre-trained Audio Encoders in the Low-Resource Condition标题：低资源条件下预训练音频编码器的研究链接：https://arxiv.org/abs/2305.17733

作者：Hao Yang,Jinming Zhao,Gholamreza Haffari,Ehsan Shareghi机构：Department of Data Science & AI, Monash University备注：INTERSPEECH 2023

【14】 StyleS2ST: Zero-shot Style Transfer for Direct Speech-to-speech Translation标题：StyleS2ST：直接语音到语音翻译的Zero-Shot风格转换链接：https://arxiv.org/abs/2305.17732

作者：Kun Song,Yi Ren,Yi Lei,Chunfeng Wang,Kun Wei,Lei Xie,Xiang Yin,Zejun Ma机构：School of Computer Science, Northwestern Polytechnical University, Xi’an, China, ByteDance备注：Accepted to Interspeech 2023

【15】 Spot keywords from very noisy and mixed speech标题：从非常嘈杂的混合语音中找出关键字链接：https://arxiv.org/abs/2305.17706

作者：Ying Shi,Dong Wang,Lantian Li,Jiqing Han,Shi Yin机构：School of Computer Science and Technology, Harbin Institute of Technology, China, Center for Speech and Language Technologies, BNRist, Tsinghua University, China, School of Artificial Intelligence, Beijing University of Posts and Telecommunications, China备注：Interspeech 2023

【16】 DPHuBERT: Joint Distillation and Pruning of Self-Supervised Speech Models标题：DPHuBERT：自监督语音模型的联合提取和剪枝链接：https://arxiv.org/abs/2305.17651

作者：Yifan Peng,Yui Sudo,Shakeel Muhammad,Shinji Watanabe机构：Carnegie Mellon University, Pittsburgh, PA, USA, Honda Research Institute Japan Co., Ltd., Saitama, Japan备注：Accepted at INTERSPEECH 2023. Code will be available at: this https URL

【17】 Translatotron 3: Speech to Speech Translation with Monolingual Data标题：Translatotron 3：单语言数据的语音到语音翻译链接：https://arxiv.org/abs/2305.17547

作者：Eliya Nachmani,Alon Levkovitch,Yifan Ding,Chulayutsh Asawaroengchai,Heiga Zen,Michelle Tadmor Ramanovich机构：Google Research,Google DeepMind

【18】 Modality-Independent Teachers Meet Weakly-Supervised Audio-Visual Event Parser标题：情态独立教师遇到弱监督视听事件解析器链接：https://arxiv.org/abs/2305.17343

作者：Yung-Hsuan Lai,Yen-Chun Chen,Yu-Chiang Frank Wang机构：National Taiwan University ,Microsoft ,NVIDIA

【19】 MT-SLVR: Multi-Task Self-Supervised Learning for Transformation In(Variant) Representations标题：MT-SLVR：用于变换(变体)表示的多任务自监督学习链接：https://arxiv.org/abs/2305.17191

作者：Calum Heggan,Tim Hospedales,Sam Budgett,Mehrdad Yaghoobi机构： School Of Engineering, University of Edinburgh, Scotland, School Of Informatics, University of Edinburgh, Scotland, Thales UK RTI备注：Last author version accepted to InterSpeech23. 5 pages

【20】 A Hierarchical Context-aware Modeling Approach for Multi-aspect and Multi-granular Pronunciation Assessment标题：多方面多粒度发音评估的层次化上下文感知建模方法链接：https://arxiv.org/abs/2305.18146

作者：Fu-An Chao,Tien-Hong Lo,Tzu-I Wu,Yao-Ting Sung,Berlin Chen机构：Research Center for Psychological and Educational Testing, National Taiwan Normal University, Department of Computer Science and Information Engineering, National Taiwan Normal University备注：Accepted to Interspeech 2023

【21】 An Experimental Review of Speaker Diarization methods with application to Two-Speaker Conversational Telephone Speech recordings标题：说话人二元化方法在双人通话电话录音中应用的实验研究链接：https://arxiv.org/abs/2305.18074

作者：Luca Serafini,Samuele Cornell,Giovanni Morrone,Enrico Zovato,Alessio Brutti,Stefano Squartini机构：Ancona, Italy, PerVoice S.p.A., Trento, Italy, Fondazione Bruno Kessler, Trento, Italy备注：52 pages, 10 figures

【22】 Automatic Evaluation of Turn-taking Cues in Conversational Speech Synthesis标题：会话语音合成中话轮转换提示的自动评价链接：https://arxiv.org/abs/2305.17971

作者：Erik Ekstedt,Siyang Wang,Éva Székely,Joakim Gustafson,Gabriel Skantze机构：KTH, Royal Institute of Technology, Stockholm, Sweden备注：Accepted at INTERSPEECH 2023, 5 pages, 2 figures, 4 tables

【23】 Stochastic Pitch Prediction Improves the Diversity and Naturalness of Speech in Glow-TTS标题：随机基音预测提高了Glow-TTS中语音的多样性和自然度链接：https://arxiv.org/abs/2305.17724

作者：Sewade Ogun,Vincent Colotte,Emmanuel Vincent机构：Universit´e de Lorraine, CNRS, Inria, LORIA, F-, Nancy, France备注：5 pages with 3 figures, InterSpeech 2023

【24】 Adapting Language-Audio Models as Few-Shot Audio Learners标题：把语言--音频模型改编成短小的音频学习者链接：https://arxiv.org/abs/2305.17719

作者：Jinhua Liang,Xubo Liu,Haohe Liu,Huy Phan,Emmanouil Benetos,Mark D. Plumbley,Wenwu Wang机构：⋆ Centre for Digital Music, Queen Mary University of London, UK, ∗ Centre for Vision, Speech and Signal Processing (CVSSP), † Amazon Alexa ‡ The Alan Turing Institute, UK

【25】 One-Step Knowledge Distillation and Fine-Tuning in Using Large Pre-Trained Self-Supervised Learning Models for Speaker Verification标题：用于说话人确认的大型预训练自监督学习模型的一步知识提取和精调链接：https://arxiv.org/abs/2305.17394

作者：Jungwoo Heo,Chan-yeong Lim,Ju-ho Kim,Hyun-seo Shin,Ha-Jin Yu机构：School of Computer Science, University of Seoul

eess.AS音频处理

【1】 A Hierarchical Context-aware Modeling Approach for Multi-aspect and Multi-granular Pronunciation Assessment标题：多方面多粒度发音评估的层次化上下文感知建模方法链接：https://arxiv.org/abs/2305.18146

【2】 An Experimental Review of Speaker Diarization methods with application to Two-Speaker Conversational Telephone Speech recordings标题：说话人二元化方法在双人通话电话录音中应用的实验研究链接：https://arxiv.org/abs/2305.18074

【3】 Automatic Evaluation of Turn-taking Cues in Conversational Speech Synthesis标题：会话语音合成中话轮转换提示的自动评价链接：https://arxiv.org/abs/2305.17971

【4】 Stochastic Pitch Prediction Improves the Diversity and Naturalness of Speech in Glow-TTS标题：随机基音预测提高了Glow-TTS中语音的多样性和自然度链接：https://arxiv.org/abs/2305.17724

作者：Sewade Ogun,Vincent Colotte,Emmanuel Vincent机构：Universit´e de Lorraine, CNRS, Inria, LORIA, F-, Nancy, France备注：5 pages with 3 figures, InterSpeech 2023

【5】 Adapting Language-Audio Models as Few-Shot Audio Learners标题：把语言--音频模型改编成短小的音频学习者链接：https://arxiv.org/abs/2305.17719

【6】 Creating Personalized Synthetic Voices from Post-Glossectomy Speech with Guided Diffusion Models标题：利用导引扩散模型从舌癌术后语音生成个性化合成语音链接：https://arxiv.org/abs/2305.17436

作者：Yusheng Tian,Guangyan Zhang,Tan Lee机构：Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong SAR备注：submitted to INTERSPEECH 2023

【7】 One-Step Knowledge Distillation and Fine-Tuning in Using Large Pre-Trained Self-Supervised Learning Models for Speaker Verification标题：用于说话人确认的大型预训练自监督学习模型的一步知识提取和精调链接：https://arxiv.org/abs/2305.17394

作者：Jungwoo Heo,Chan-yeong Lim,Ju-ho Kim,Hyun-seo Shin,Ha-Jin Yu机构：School of Computer Science, University of Seoul

【8】 CommonAccent: Exploring Large Acoustic Pretrained Models for Accent Classification Based on Common Voice标题：CommonAccent：探索基于普通语音的大型声学预训练模型用于口音分类链接：https://arxiv.org/abs/2305.18283

作者：Juan Zuluaga-Gomez,Sara Ahmed,Danielius Visockas,Cem Subakan机构：† Idiap Research Institute, Switzerland ‡ Ecole Polytechnique Federale de Lausanne, Switzerland, Sketch Recognition Lab, Texas A&M University, USA, § Vilnius Gediminas Technical University, Vilnius, Lithuania备注：To appear in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2023

【9】 HyperConformer: Multi-head HyperMixer for Efficient Speech Recognition标题：HyperConformer：用于高效语音识别的多头超级混合器链接：https://arxiv.org/abs/2305.18281

作者：Florian Mai,Juan Zuluaga-Gomez,Titouan Parcollet,Petr Motlicek机构：† Idiap Research Institute, Martigny, Switzerland, ‡ Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland, ¶ University of Cambridge, United Kingdom, § Brno University of Technology, Brno, Czech Republic备注：Florian Mai and Juan Zuluaga-Gomez contributed equally. To appear in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2023

【10】 Exploration of Efficient End-to-End ASR using Discretized Input from Self-Supervised Learning标题：利用离散化的自监督学习输入探索高效的端到端ASR链接：https://arxiv.org/abs/2305.18108

作者：Xuankai Chang,Brian Yan,Yuya Fujita,Takashi Maekaku,Shinji Watanabe机构：Carnegie Mellon University, PA, USA, Yahoo Japan Corporation, Japan备注：Accepted at INTERSPEECH 2023

【11】 Improving Textless Spoken Language Understanding with Discrete Units as Intermediate Target标题：以离散单元为中介目标提高无文本口语理解能力链接：https://arxiv.org/abs/2305.18096

作者：Guan-Wei Wu,Guan-Ting Lin,Shang-Wen Li,Hung-yi Lee机构：National Taiwan University, Taiwan, Meta AI, USA备注：Accepted by interspeech 2023

【12】 Few-shot Class-incremental Audio Classification Using Adaptively-refined Prototypes标题：基于自适应细化原型的Few-Shot类增量音频分类链接：https://arxiv.org/abs/2305.18045

【13】 ADAPTERMIX: Exploring the Efficacy of Mixture of Adapters for Low-Resource TTS Adaptation标题：ADAPTERMIX：探索混合适配器对低资源TTS适应的有效性链接：https://arxiv.org/abs/2305.18028

【14】 Can We Trust Explainable AI Methods on ASR? An Evaluation on Phoneme Recognition标题：我们能在ASR上信任可解释的人工智能方法吗？音素识别研究述评链接：https://arxiv.org/abs/2305.18011

作者：Xiaoliang Wu,Peter Bell,Ajitha Rajan机构：School of Informatics, University of Edinburgh

【15】 Multi-Scale Attention for Audio Question Answering标题：音频问答中的多尺度注意链接：https://arxiv.org/abs/2305.17993

【16】 speech and noise dual-stream spectrogram refine network with speech distortion loss for robust speech recognition标题：用于稳健语音识别的语音和噪声双流谱图精化语音失真损失网络链接：https://arxiv.org/abs/2305.17860

【17】 Retraining-free Customized ASR for Enharmonic Words Based on a Named-Entity-Aware Model and Phoneme Similarity Estimation标题：基于命名实体感知模型和音素相似度估计的免再训练非调和词定制ASR链接：https://arxiv.org/abs/2305.17846

【18】 Streaming Audio Transformers for Online Audio Tagging标题：用于在线音频标记的流音频转换器链接：https://arxiv.org/abs/2305.17834

作者：Heinrich Dinkel,Zhiyong Yan,Yongqing Wang,Junbo Zhang,Yujun Wang机构：Xiaomi Corporation, Beijing, China

【19】 RASR2: The RWTH ASR Toolkit for Generic Sequence-to-sequence Speech Recognition标题：RASR2：用于通用序列到序列语音识别的RWTH ASR工具包链接：https://arxiv.org/abs/2305.17782

【20】 CAPTDURE: Captioned Sound Dataset of Single Sources标题：CAPTDURE：单源字幕声音数据集链接：https://arxiv.org/abs/2305.17758

【21】 Bayesian inference and neural estimation of acoustic wave propagation标题：声波传播的贝叶斯推断和神经估计链接：https://arxiv.org/abs/2305.17749

作者：Yongchao Huang,Yuhang He,Hong Ge机构： Department of Computer Science, University of Oxford‡hg 3 4 4

【22】 Range-Based Equal Error Rate for Spoof Localization标题：基于测距的等错误率欺骗定位方法链接：https://arxiv.org/abs/2305.17739

【23】 Investigating Pre-trained Audio Encoders in the Low-Resource Condition标题：低资源条件下预训练音频编码器的研究链接：https://arxiv.org/abs/2305.17733

作者：Hao Yang,Jinming Zhao,Gholamreza Haffari,Ehsan Shareghi机构：Department of Data Science & AI, Monash University备注：INTERSPEECH 2023

【24】 StyleS2ST: Zero-shot Style Transfer for Direct Speech-to-speech Translation标题：StyleS2ST：直接语音到语音翻译的Zero-Shot风格转换链接：https://arxiv.org/abs/2305.17732

【25】 Spot keywords from very noisy and mixed speech标题：从非常嘈杂的混合语音中找出关键字链接：https://arxiv.org/abs/2305.17706

【26】 DPHuBERT: Joint Distillation and Pruning of Self-Supervised Speech Models标题：DPHuBERT：自监督语音模型的联合提取和剪枝链接：https://arxiv.org/abs/2305.17651

【27】 Translatotron 3: Speech to Speech Translation with Monolingual Data标题：Translatotron 3：单语言数据的语音到语音翻译链接：https://arxiv.org/abs/2305.17547

作者：Eliya Nachmani,Alon Levkovitch,Yifan Ding,Chulayutsh Asawaroengchai,Heiga Zen,Michelle Tadmor Ramanovich机构：Google Research,Google DeepMind

【28】 CIF-PT: Bridging Speech and Text Representations for Spoken Language Understanding via Continuous Integrate-and-Fire Pre-Training标题：CIF-PT：通过持续整合和激发预训练为口语理解搭建语音和文本表示之间的桥梁链接：https://arxiv.org/abs/2305.17499

作者：Linhao Dong,Zhecheng An,Peihao Wu,Jun Zhang,Lu Lu,Zejun Ma机构：ByteDance AI Lab备注：Accepted by ACL 2023 Findings

【29】 Modality-Independent Teachers Meet Weakly-Supervised Audio-Visual Event Parser标题：情态独立教师遇到弱监督视听事件解析器链接：https://arxiv.org/abs/2305.17343

作者：Yung-Hsuan Lai,Yen-Chun Chen,Yu-Chiang Frank Wang机构：National Taiwan University ,Microsoft ,NVIDIA

【30】 MT-SLVR: Multi-Task Self-Supervised Learning for Transformation In(Variant) Representations标题：MT-SLVR：用于变换(变体)表示的多任务自监督学习链接：https://arxiv.org/abs/2305.17191

【31】 Visually grounded few-shot word acquisition with fewer shots标题：用较少的镜头获得视觉接地的Few-Shot单词链接：https://arxiv.org/abs/2305.15937

作者：Leanne Nortje,Benjamin van Niekerk,Herman Kamper机构：MediaLab, E&E Engineering, Stellenbosch University, South Africa备注：Accepted at Interspeech 2023

机器翻译，仅供参考

语音/音频处理学术速递[5.30]

评论 0