L2-GEN: A Neural Phoneme Paraphrasing Approach to L2 Speech Synthesis for Mispronunciation Diagnosis.
Daniel ZhangAshwinkumar GanesanSarah CampbellDaniel KorzekwaPublished in: INTERSPEECH (2022)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- prosodic features
- network architecture
- neural network
- context sensitive
- vocal tract
- model based diagnosis
- speech corpus
- automatic diagnosis
- fault diagnosis
- context specific
- bio inspired
- fault detection
- medical diagnosis
- pattern recognition
- information retrieval
- neural fuzzy
- question answering
- model based reasoning
- associative memory
- biologically plausible
- subband
- diagnostic reasoning
- hebbian learning
- hidden markov models
- neural model
- clinically relevant
- expert systems
- noisy environments
- speech signal