Voice Engine by OpenAI requires a single 15-second audio sample to generate natural-sounding speech closely resembling the original speaker.
Partners using Voice Engine must follow usage policies, including obtaining consent from the original speaker, disclosing AI usage, and watermarking generated audio.
OpenAI recommends voice authentication for ethical use of synthetic voices and advocates phasing out voice-based authentication for sensitive tasks.
OpenAI expressed uncertainty on the widespread release of Voice Engine, aiming to provoke discussions on responsible synthetic voice deployment.
[
add
]
[
|
|
...
]