读论文
文章平均质量分 92
人工智能领域论文阅读
让我看看谁在学习
学无止境
展开
-
Automatic Lip-reading with Hierarchical Pyramidal Convolution and Self-Attention for Image Sequences
基于分层金字塔卷积和自注意力的无单词边界图像序列自动唇读原创 2022-07-30 19:44:27 · 453 阅读 · 1 评论 -
读Training Strategies for Improved Lip-Reading论文
改善唇读的训练策略原创 2022-07-08 21:03:56 · 1596 阅读 · 0 评论 -
读Leveraging Unimodal Self-Supervised Learning for Multimodal AVSR论文
标题:利用单模态自监督学习进行多模态视听语音识别论文:https://arxiv.org/pdf/2203.07996v2.pdf代码:https://github.com/lumia-group/leveraging-self-supervised-learning-for-avsr关键词:audio-visual speech recognition (A VSR)视听语音识别、unimodal data单模态数据、self-supervised learning自监督学习、CTC和Seq2原创 2022-05-31 17:45:37 · 913 阅读 · 4 评论 -
读Hearing Lips:Improving Lip Reading by Distilling Speech Recognizers论文
论文:https://arxiv.org/pdf/1911.11502.pdf代码:无标题:听唇:通过蒸馏语音识别器改善唇读关键词:多模态、语音唇读LIBS、CMLR中文数据集、Lip by Speech (LIBS)、CSSMCM、attention-based sequence-to-sequence model[sos] => 句子起始标识符、[eos] => 句子结束标识符和 [pad] => 补全字符、word embedding:通俗的翻译可以认为是单词嵌入原创 2022-05-26 17:49:46 · 927 阅读 · 3 评论