Voice Cloning takes a very small sample of your voice and creates a synthetic model of your voice.
Two Minute Papers explains a new technique that only needs 5 seconds of your real voice (it infers the rest).
Original “Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis” paper: https://arxiv.org/abs/1806.04558
Medium Article on the same subject: https://medium.com/syncedreview/clone-a-voice-in-five-seconds-with-this-ai-toolbox-f3f116b11281
No comments:
Post a Comment