From Vox Media.: 2024-03-29 19:10:15

OpenAI has introduced Voice Engine, a text-to-voice platform that creates synthetic voices from a 15-second clip. Companies like Age of Learning, HeyGen, Dimagi, Livox, and Lifespan have access. Technology can generate voice-over content and respond to students. Voice Engine was developed in late 2022 and powers ChatGPT’s Read Aloud feature.

AI text-to-audio generation is evolving with OpenAI’s Voice Engine. Partners must follow usage policies, including not impersonating individuals without consent. Voice generation should disclose AI origin and not allow users to create voices. Watermarking is added to trace audio clips. OpenAI recommends phasing out voice-based authentication and developing AI content tracking systems.



Read more at Vox Media.: OpenAI’s voice cloning AI model only needs a 15-second sample to work