Articulatory Copy Synthesis Based on the Speech Synthesizer VocalTractLab and Convolutional Recurrent Neural Networks.
Yingming GaoPeter BirkholzYa LiPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2024)
Keyphrases
- recurrent neural networks
- speech synthesis
- text to speech
- vocal tract
- speech recognition
- speech signal
- multi stream
- feed forward
- neural network
- recurrent networks
- echo state networks
- feedforward neural networks
- complex valued
- automatic speech recognition
- hidden layer
- reservoir computing
- acoustic features
- neural model
- formant frequencies
- hidden markov models
- long short term memory
- cascade correlation
- nonlinear dynamic systems
- artificial neural networks
- audio visual
- restricted boltzmann machine
- facial animation
- chaotic time series
- deep learning
- neural models
- word processing
- linear prediction
- language model
- long term