Phonet: A Tool Based on Gated Recurrent Neural Networks to Extract Phonological Posteriors from Speech.
Juan Camilo Vásquez-CorreaPhilipp KlumppJuan Rafael Orozco-ArroyaveElmar NöthPublished in: INTERSPEECH (2019)
Keyphrases
- recurrent neural networks
- neural network
- echo state networks
- recurrent networks
- feed forward
- feedforward neural networks
- complex valued
- speech recognition
- long short term memory
- neural model
- artificial neural networks
- reservoir computing
- speech signal
- text to speech
- cascade correlation
- text to speech synthesis
- nonlinear dynamic systems
- audio visual
- posterior probability
- probabilistic model
- expert systems