Two years ago, technologists at OpenAI published a blog post describing how the company developed the ability to replicate human voices using powerful artificial intelligence technology. The new software was so advanced, the researchers said, that OpenAI decided not to release it to the public out of an abundance of caution.
While the company’s position hasn’t changed, it hasn’t stopped OpenAI from continuing to work on the technology. This year, OpenAI quietly bought Weights.gg, a small startup that offered artificial intelligence tools to create clones of people’s voices, two people familiar with the deal said on condition of anonymity because they were not authorized to speak publicly.
Terms of the transaction, which were not publicly announced, could not be determined. People familiar with the acquisition said OpenAI bought both the company’s small team of employees and its intellectual property. Weights.gg has announced that it is discontinuing its services in March.
Weights.gg operated as a kind of social network for creating and sharing artificial intelligence algorithms that could do things like clone voices using the company’s free consumer app, called Replay.
One YouTube user showed how he used Weights.gg technology to clone actor Samuel L. Jackson’s voice. Some of the best voice models in the company’s repository have included clones of Taylor Swift, Kanye West, and members of the popular K-pop band Blackpink.
Weights.gg tools have also been used to clone other copyrighted voices, including Bugs Bunny and Daffy Duck. The Weights.gg library has also featured the voices of prominent political figures, including President Trump and former President Joseph R. Biden Jr.
Some high-profile figures, including Mr Jackson, have spoken out against having their voices cloned using the technology. Last month, Mrs. Swift filed a series of applications to the US Patent and Trademark Office to trademark her voice and likeness.
Weights.gg employed half a dozen people and raised roughly $4 million in venture capital, according to data compiled by research firm PitchBook.
OpenAI has faced copyright issues in the past. Last year, the company released Sora, a smartphone app that allowed people to instantly create videos of copyrighted characters without their permission. The company quickly ran into opposition from Hollywood before making deals to use them.
(The New York Times has sued OpenAI and Microsoft, alleging copyright infringement of news content related to AI systems. Both companies have denied the suit’s claims.)
It’s unclear what OpenAI plans to do with the Weights.gg team and technology. But the Weights.gg staff has broken up to work in teams across different parts of the company, the two people said. They are unlikely to release a product similar to Weights.gg, they said.
In recent months, OpenAI has scaled back some of its ambitions to focus on revenue-generating products as it looks set to begin trading as a public company by the end of this year. OpenAI shut down the Sora video app this year. She also tried to repair her ties to Hollywood poaching the “celebrity whisperer” of Instagram Charles Porch to mend frayed relationships.
Instead, it focused on incorporating its voice technology into other parts of the company. This month, OpenAI released information about how third-party developers can use the company’s application programming interface, or API, to incorporate OpenAI’s voice technology into external applications and services. These uses, for which developers would pay, could include providing real-time voice translation services or interacting with “agents” using voice commands.
OpenAI also released its ChatGPT at Apple Car Play appwhich allows people to give voice commands to the chatbot while driving. This feature has improved over time thanks to OpenAI’s investments in voice technology.
According to the company’s 2024 blog post, OpenAI appears to have no immediate plans to release its voice cloning technology outside of a limited group of partners. And in others blog postsOpenAI has said it supports increased security practices around voice cloning technologies like the ones it created.
