David Bridges
In case you’ve been dreaming about spending your summer time whispering candy nothings into the digital ears of one of many seductive ChatGPT voice assistants that OpenAI confirmed off final month, you’ll need to dream a bit longer. On Tuesday, the corporate introduced that its “superior Voice Mode” characteristic wants extra time within the oven “to succeed in our bar to launch.” The characteristic will probably be out there to a small group of customers to collect suggestions, after which launch to all paying ChatGPT prospects within the fall.
“We’re enhancing the mannequin’s potential to detect and refuse sure content material,” OpenAI posted on X. “We’re additionally engaged on enhancing the consumer expertise and making ready our infrastructure to scale to tens of millions whereas sustaining real-time responses.”
We’re sharing an replace on the superior Voice Mode we demoed throughout our Spring Replace, which we stay very enthusiastic about:
We had deliberate to begin rolling this out in alpha to a small group of ChatGPT Plus customers in late June, however want yet another month to succeed in our bar to launch.…
— OpenAI (@OpenAI) June 25, 2024
Voices have been part of ChatGPT since 2023. However final month, OpenAI demoed an upgraded model that sounded so humanlike it drew comparisons with Samantha, the seductive voice assistant within the 2013 film Her, performed by Scarlett Johansson. Weeks after the presentation, the actress accused OpenAI of copying her voice regardless of denying permission.
OpenAI stated it is nonetheless determining when the brand new voices (minus the Johansson soundalike) will roll out to paying customers this fall. One other characteristic that lets the voice assistants use your cellphone’s digital camera to know the world round you has additionally been delayed till that point. “Precise timelines rely on assembly our excessive security and reliability bar,” the corporate stated.
Daniel Mercer
