All Tools
V
OtherFreeOpen Source
VOICECRAFT
VoiceCraft is a token infilling neural codec language model for zero-shot speech editing and text-to-speech
ABOUT
Editing speech recordings and synthesizing new speech in a specific voice traditionally requires extensive training data for each target speaker and complex voice cloning pipelines. VoiceCraft solves this with a token infilling approach that enables zero-shot speech editing and TTS — modifying spoken words in existing recordings or generating new speech in an unseen voice using just seconds of reference audio.
INTEGRATION GUIDE
1. Edit spoken words in existing audio recordings without rerecording
2. Generate natural speech in a target voice from just seconds of reference
3. Create personalized voice interfaces and audiobook narration with voice cloning