同步公衆號(arXiv每日學術速遞)
【1】 Bayesian Restoration of Audio Degraded by Low-Frequency Pulses Modeled via Gaussian Process
標題:用高斯過程建模的低頻脈衝退化音頻的貝葉斯恢復
作者: Hugo Tremonte de Carvalho, Luiz Wagner Pereira Biscainho
鏈接:https://arxiv.org/abs/2005.14181
【2】 The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Testing Framework, and Challenge Results
標題:InterSpeech 2020深度噪聲抑制挑戰:數據集、主觀測試框架和挑戰結果
作者: Chandan K. A. Reddy, Johannes Gehrke
備註:Interspeech 2020. arXiv admin note: substantial text overlap with arXiv:2001.08662
鏈接:https://arxiv.org/abs/2005.13981
【3】 When Can Self-Attention Be Replaced by Feed Forward Layers?
標題:什麼時候可以用前饋層代替自我關注?
作者: Shucong Zhang, Steve Renals
鏈接:https://arxiv.org/abs/2005.13895
【4】 Speech-to-Singing Conversion based on Boundary Equilibrium GAN
標題:基於邊界平衡GaN的語音-演唱轉換
作者: Da-Yi Wu, Yi-Hsuan Yang
鏈接:https://arxiv.org/abs/2005.13835
【5】 Subword RNNLM Approximations for Out-Of-Vocabulary Keyword Search
標題:用於詞彙表外關鍵詞搜索的子詞RNNLM近似
作者: Mittul Singh, Mikko Kurimo
備註:INTERSPEECH 2019
鏈接:https://arxiv.org/abs/2005.13827
【6】 DeepSonar: Towards Effective and Robust Detection of AI-Synthesized Fake Voices
標題:DeepSonar:促進有效和穩健檢測AI合成的假聲
作者: Run Wang, Yang Liu
鏈接:https://arxiv.org/abs/2005.13770
【7】 Unsupervised Audio Source Separation using Generative Priors
標題:基於生成先驗的無監督音頻源分離
作者: Vivek Narayanaswamy, Andreas Spanias
鏈接:https://arxiv.org/abs/2005.13769
【8】 Phone Features Improve Speech Translation
標題:電話功能改善語音翻譯
作者: Elizabeth Salesky, Alan W Black
備註:Accepted to ACL2020
鏈接:https://arxiv.org/abs/2005.13681
【9】 Modality Dropout for Improved Performance-driven Talking Faces
標題:改進的性能驅動的會說話面孔的模態丟棄(Modality Dropout)
作者: Ahmed Hussen Abdelaziz, Sachin Kajareker
鏈接:https://arxiv.org/abs/2005.1361