Interpreting DNN Output Layer Activations: A Strategy to Cope with Unseen Data in Speech Recognition.
Vikramjit MitraHoracio FrancoPublished in: ICASSP (2018)
Keyphrases
- speech recognition
- unseen data
- hidden markov models
- speech signal
- cross validation
- pattern recognition
- language model
- output layer
- hidden layer
- test data
- speech recognition systems
- training data
- classification rules
- training set
- feedforward neural networks
- feed forward
- rbf neural network
- input output
- radial basis function
- neural network
- model selection
- artificial neural networks
- face recognition