Few-Shot Speaker Identification Using Depthwise Separable Convolutional Network with Channel Attention.
Yanxiong LiWucheng WangHao ChenWenchang CaoWei LiQianhua HePublished in: CoRR (2022)
Keyphrases
- pattern recognition
- speaker identification
- speech recognition
- feature extraction
- convolutional network
- gaussian mixture model
- convolutional neural networks
- speech signal
- machine learning
- noisy environments
- feature extractor
- broadcast news
- focus of attention
- visual attention
- video content
- object recognition
- language model
- coarse to fine