An end-to-end spoofing countermeasure for automatic speaker verification using evolving recurrent neural networks.
Giacomo ValentiHéctor DelgadoMassimiliano TodiscoNicholas W. D. EvansLaurent PilatiPublished in: Odyssey (2018)
Keyphrases
- end to end
- recurrent neural networks
- speaker verification
- recurrent networks
- feed forward
- echo state networks
- neural network
- noisy environments
- congestion control
- artificial neural networks
- audio visual
- face verification
- emotion recognition
- application layer
- information fusion
- language identification
- non stationary
- artificial intelligence
- learning algorithm