Generative De-Quantization for Neural Speech Codec via Latent Diffusion.
Haici YangInseon JangMinje KimPublished in: CoRR (2023)
Keyphrases
- speech recognition
- network architecture
- neural network
- generative model
- anisotropic diffusion
- latent variables
- speech signal
- diffusion process
- speech synthesis
- motion estimation
- video coding
- quantization error
- video codec
- text to speech
- audio visual
- neural model
- generative process
- automatic speech recognition
- semi supervised
- spoken language
- gaussian process latent variable models
- probabilistic topic models
- broadcast news
- distributed video coding
- diffusion model
- information diffusion
- coding method
- vector quantization
- unsupervised learning
- hidden markov models