Alan Baade

self-supervised learning (2) voice conversion (1) autoregressive generation (1) visual correspondence (1) language model (1) masked autoencoder (1) encoder-decoder architecture (1) neural codec (1) speaker disentanglement (1) semantic unit (1) textless speech synthesis (1) neural codec language model (1) discrete semantic unit (1) classifier free guidance (1) audio spectrogram transformer (1) cross-view correspondence (1) predictive cycle consistency (1) vision transformer (1) correspondence refinement (1)