Residual convolutional neural network with attentive feature pooling for end-to-end language identification from short-duration speech.
João MonteiroMd. Jahangir AlamTiago H. FalkPublished in: Comput. Speech Lang. (2019)
Keyphrases
- end to end
- language identification
- convolutional neural network
- speaker identification
- multi lingual
- english text
- speech recognition
- speech signal
- noisy environments
- face detection
- gaussian mixture model
- document images
- feature extraction
- broadcast news
- neural network
- visual attention
- congestion control
- audio visual
- feature selection
- feature vectors
- multi modal
- feature set
- hidden markov models