Deep Reinforcement Learning for Articulatory Synthesis in a Vowel-to-Vowel Imitation Task.
Denis ShitovElena PirogovaTadeusz A. WysockiMargaret LechPublished in: Sensors (2023)
Keyphrases
- reinforcement learning
- vocal tract
- formant frequencies
- speech synthesis
- speech signal
- speech recognition
- prosodic features
- function approximation
- reinforcement learning algorithms
- fundamental frequency
- state space
- model free
- machine learning
- reinforcement learning methods
- linear prediction
- automatic speech recognition
- markov decision processes
- pattern recognition
- learning algorithm
- image acquisition
- optimal policy
- language model
- dynamic programming
- imitation learning
- multi agent
- policy search
- speaker independent
- genetic algorithm
- real time