DelightfulTTS 2: End-to-End Speech Synthesis with Adversarial Vector-Quantized Auto-Encoders.
Yanqing LiuRuiqing XueLei HeXu TanSheng ZhaoPublished in: CoRR (2022)
Keyphrases
- end to end
- speech synthesis
- vector quantized
- vector quantization
- speech recognition
- image compression
- progressive transmission
- text to speech
- visual words
- video compression
- image coding
- congestion control
- hidden markov models
- pattern recognition
- bayesian networks
- bag of words
- image classification
- error propagation
- co occurrence
- wavelet transform
- bitstream
- compression algorithm
- multiresolution