Abstractive headline generation for spoken content by attentive recurrent neural networks with ASR error modeling.
Lang-Chi YuHung-yi LeeLin-Shan LeePublished in: SLT (2016)
Keyphrases
- recurrent neural networks
- echo state networks
- neural network
- reservoir computing
- recurrent networks
- feed forward
- complex valued
- artificial neural networks
- error rate
- feedforward neural networks
- speech recognition
- automatic speech recognition
- hidden layer
- neural model
- nonlinear dynamic systems
- particle swarm optimization
- real time