Segmenting acoustic signal with articulatory movement using Recurrent Neural Network for phoneme acquisition.
Hisashi KandaTetsuya OgataKazunori KomataniHiroshi G. OkunoPublished in: IROS (2008)
Keyphrases
- recurrent neural networks
- acoustic signal
- speech recognition
- acoustic signals
- speech synthesis
- neural network
- feed forward
- complex valued
- recurrent networks
- reservoir computing
- artificial neural networks
- hebbian learning
- echo state networks
- hidden layer
- automatic speech recognition
- neural model
- pattern recognition
- speech signal
- hidden markov models
- vocal tract
- feedforward neural networks
- nonlinear dynamic systems
- language model
- speaker dependent
- neural nets
- back propagation