Wangjin Zhou

self-supervised learning (1) flow matching (1) speaker verification (1) model fusion (1) acoustic model (1) disentangled representation (1) mutual information minimization (1) latent diffusion model (1) text-to-speech synthesis (1) speech-to-speech translation (1) neural vocoder (1) speech-to-text translation (1) speech quality assessment (1) text-to-audio generation (1) audio synthesis (1) mean opinion score (1) age-invariant embedding (1) dual-decoder transformer (1) cross-age verification (1) continuous normalization flow (1)