MFA-Conformer: Multi-scale Feature Aggregation Conformer for Automatic Speaker Verification.
Yang ZhangZhiqiang LvHaibin WuShanshan ZhangPengfei HuZhiyong WuHung-Yi LeeHelen MengPublished in: CoRR (2022)
Keyphrases
- speaker verification
- multiscale
- noisy environments
- speaker recognition
- prosodic features
- edge detection
- image segmentation
- feature vectors
- audio visual
- multi modal
- emotion recognition
- using artificial neural networks
- hidden markov models
- action recognition
- multilayer perceptron
- image representation
- neural network
- image classification
- feature set
- image features
- image processing