Transformer VQ-VAE for Unsupervised Unit Discovery and Speech Synthesis: ZeroSpeech 2020 Challenge.
Andros TjandraSakriani SaktiSatoshi NakamuraPublished in: CoRR (2020)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- vector quantization
- vocal tract
- image compression
- prosodic features
- data driven
- vector quantizer
- unsupervised learning
- image coding
- fractal image coding
- semi supervised
- pattern discovery
- fuzzy logic
- unsupervised manner
- knowledge discovery
- speech corpus
- power system
- machine learning
- image quality
- supervised learning
- hidden markov models