Language Identification with Deep Bottleneck Features
時間：2018.9 沒看到發佈的地址 作者：Zhanyu Ma, Hong Yu abstract 用於中文音素分類的DNN的bottleneck特徵被用於LSTM的訓練。 1. introduction SLD任務用在語音識別的前端，首先判別一句話的語種，然後喚醒對應的decode去翻譯成文本。 世界上有上千種語言，每種語言有不同的區分特徵，想要找到一種通用的，快速響應的有效SLD system
相關文章
- 1. 深度學習之Bottleneck Layer or Bottleneck Features
- 2. BN for Language Identification
- 3. Learning Transferable Features with Deep Adaptation Networks
- 4. Large-Scale and Language-Oblivious Code Authorship Identification
- 5. 【Person Re-ID】Deep-Person: Learning Discriminative Deep Features for Person Re-Identification
- 6. [c++] C Language Features
- 7. 【ReID】Learning Discriminative Features with Multiple Granularities for Person Re-identification
- 8. 翻譯「Learning Transferable Features with Deep Adaptation Networks」
- 9. Content-based image retrieval with compact deep convolutional features
- 10. Deep Reinforcement Learning with a Natural Language Action Space
- 更多相關文章...
- • RSS
元素 - RSS 教程 - • XSLT
- • YAML 入門教程
- • 爲了進字節跳動，我精選了29道Java經典算法題，帶詳細講解
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. 深度學習之Bottleneck Layer or Bottleneck Features
- 2. BN for Language Identification
- 3. Learning Transferable Features with Deep Adaptation Networks
- 4. Large-Scale and Language-Oblivious Code Authorship Identification
- 5. 【Person Re-ID】Deep-Person: Learning Discriminative Deep Features for Person Re-Identification
- 6. [c++] C Language Features
- 7. 【ReID】Learning Discriminative Features with Multiple Granularities for Person Re-identification
- 8. 翻譯「Learning Transferable Features with Deep Adaptation Networks」
- 9. Content-based image retrieval with compact deep convolutional features
- 10. Deep Reinforcement Learning with a Natural Language Action Space