Using Text and Acoustic Features in Predicting Glottal Excitation Waveforms for Parametric Speech Synthesis with Recurrent Neural Networks.
Lauri JuvelaXin WangShinji TakakiManu AiraksinenJunichi YamagishiPaavo AlkuPublished in: INTERSPEECH (2016)
Keyphrases
- recurrent neural networks
- speech synthesis
- speech signal
- speech recognition
- acoustic features
- vocal tract
- text to speech
- automatic speech recognition
- hidden markov models
- feed forward
- neural network
- language model
- echo state networks
- noisy environments
- recurrent networks
- information retrieval
- speaker identification
- pattern recognition
- text mining
- speaker verification
- artificial neural networks
- non stationary
- face recognition
- semantic information
- text data
- text documents
- music information retrieval
- back propagation
- probabilistic model
- bayesian networks
- genetic algorithm
- data mining