Meta Announces Voicebox: Clone a Voice in Any Language with 2-Sec of Audio
AI Chat: ChatGPT, AI News, Artificial Intelligence, OpenAI, Machine Learning
00:00
The Future of AI and Voice Box
A new AI model called voice box can take in up to two seconds of audio. It is able to clone a person's voice and use it for cross lingual style transfer. Voice box could give natural sounding voices to virtual assistants or non player characters on the metaverse. They could allow visually impaired people to hear written messages from friends and read them by their voices.
Play episode from 03:47
Transcript
