Prosody Correction Preserving Speaker Individuality for Chinese-Accented Japanese HMM-Based Text-to-Speech Synthesis.
Daiki SekizawaShinnosuke TakamichiHiroshi SaruwatariPublished in: IEICE Trans. Inf. Syst. (2019)
Keyphrases
- text to speech synthesis
- text to speech
- prosodic features
- speech synthesis
- japanese language
- native speakers
- chinese characters
- hidden markov models
- synthesized speech
- error analysis
- word processing
- speech recognition
- error correction
- automatic speech recognition
- multi modal
- dependency parser
- speaker verification
- image details
- word segmentation
- audio visual
- language learning
- distortion correction
- chinese text
- foreign language
- neural network