
VoxGenesis introduces an unsupervised speech synthesis framework that discovers a latent speaker manifold and enables voice editing without supervision. By transforming a Gaussian distribution into speech distributions conditioned by semantic tokens, VoxGenesis disentangles speaker characteristics from content information.


coremsg

Unsupervised-Discovery-of-Latent-Speaker-Manifold-for-Speech-Synthesis-VoxGenesis


Unsupervised Discovery of Latent Speaker Manifold for Speech Synthesis: VoxGenesis


title_rewrite


VoxGenesisは、監督なしの音声合成フレームワークであり、潜在的な話者マニフォールドを発見し、意味のある音声編集方向を探索します。


voxgenesis-unsupervised-discovery-of-latent-speaker-manifold-for-speech-synthesis


VoxGenesis: Unsupervised Discovery of Latent Speaker Manifold for Speech Synthesis
