Last summer, as they went to a doctor's appointment near their home in Manhattan, Paul Sky Lehrman and Lena Sage talked about the rise of artificial intelligence and the threat to the livelihoods of writers, actors and other entertainment professionals. I listened to a podcast.
This topic was especially important for young married couples. They made their living as voice actors, and AI technologies began to produce voices that sounded like the real thing.
But the podcast had an unexpected twist. To highlight the threat of AI, the host conducted a lengthy interview with a talking chatbot named Poe. It sounded just like Mr. Luhrmann.
“He was interviewing my voice about the dangers of AI and the damage it could do to the entertainment industry,” Mr. Lehrman said. “We pulled over and sat there in utter disbelief, trying to figure out what had just happened and what we should do.”
Mr. Lehrman and Ms. Sage are now suing the company that made the bot's voice. They claim that Luvo, a startup in Berkeley, California, illegally used their voice recordings to create technology that could compete with their voice work. After hearing a clone of Mr. Lehrman's voice on a podcast, the couple discovered that Lowe had also cloned Ms. Sage's voice.
The pair joins a growing number of artists, publishers, computer programmers and other creators who have filed lawsuits against the makers of AI technologies, arguing that the companies used their work to create such tools without permission. I used to eventually replace them in the job market. (The New York Times sued two companies, OpenAI and its partner, Microsoft, in December, accusing them of using its copyrighted news articles to build their online chatbots. .)
In their lawsuit, filed Thursday in Manhattan federal court, the couple said anonymous employees of Lau paid them for a handful of audio clips in 2019 and 2020, without disclosing how the clips were to be used. will
They say Lovo, which was founded in 2019, is violating federal trademark law and several state privacy laws by promoting clones of its voices. The suit seeks class action status, with Mr. Lehrman and Ms. Sage inviting other voice actors to join.
“We don't know how many other people have been affected,” said his attorney, Steve Cohen.
David Case, an attorney representing the company, said Luove denies the claims in the lawsuit. He added that if all the people providing voice recordings to Luvo give their consent, there is no problem.
The company's chief executive, Tom Lee, said in a podcast episode last year that Lowe now offers a revenue-sharing program that allows voice actors to help the company create voice clones of themselves. And get a cut of the money earned by these clones.
Jeffrey Bennett, general counsel for SAG-AFTRA, the labor union that represents 160,000 media professionals worldwide, said the lawsuit appears to be the first of its kind.
“This suit will show people — especially technology companies — that there are rights to your voice, that there's a whole group of people out there who are making a living using their voice,” he said.
In 2019, Mr. Lehrman and Ms. Sage were promoting themselves as voice actors on Fiverr, a website where freelance professionals can advertise their work. Through this online marketplace, he was often asked to provide voice work for commercials, radio commercials, online videos, video games and other media.
That year, Ms. Sage was contacted by an anonymous person who paid her $400 to record several radio scripts and explained that the recordings would not be used for public purposes, according to the lawsuit. .
“These are test scripts for radio commercials,” the anonymous person said, according to the suit. “They will not be displayed externally, and will only be used internally, so no rights are required.” “
Seven months later, another unidentified person approached Mr. Lehrman about a similar job. Mr. Lehrman, who also works as a television and film actor, asked how the clips would be used. The man said multiple times that they would be used only for research and academic purposes, according to correspondence in the suit. Mr. Lehrman was paid $1,200. (She provided a longer recording than Ms. Sage.)
In April 2022, Mr. Lehrman discovered a YouTube video about the war in Ukraine narrated by a voice that sounded like him.
“This is my voice talking about weapons in the Ukraine-Russia conflict,” he said. “I'm ghostly white—goose bumps on my arms. I knew I'd never said those words in that order.
For months, he and Ms. Sage struggled to understand what had happened. They hired a lawyer to help find out who made the YouTube video and how Mr. Lehrman's voice was recreated. But the owner of the YouTube channel seemed to be based in Indonesia, and they had no way of finding the person.
Then they listened to the podcast on their way to the doctor's office. Through the podcast, “Deadline Strike Talk,” they were able to identify the source of the clone of Mr. Lehrman's voice. A professor at the Massachusetts Institute of Technology put together the chatbot using Lovo's voice synthesis technology.
Ms. Sage also found an online video of the company pitching its voice technology to investors during an event in Berkeley in early 2020. In the video, a Luove executive showed a synthetic version of Ms. Sage's voice and compared it to a recording of her real voice. Both played with the image of a woman she was not.
“I was in their pitch video to raise money,” Ms Sage said. The company has since raised more than $7 million and claims more than two million users worldwide.
Mr. Lehrman and Ms. Sage also discovered that Lovo was promoting their and Mr. Lehrman's voice clones on its website. After sending the company a cease-and-desist letter, the company said it had removed their voice clones from the site. But Mr. Lehrman and Ms. Sage argued that the software to create these voice clones had already been downloaded and could still be used by countless numbers of the company's customers.
Mr. Lehrman also questioned whether the company used the pair's voices, along with many others, to develop the underlying technology that drives its voice cloning system. Voice synthesizers often learn their skills by analyzing thousands of hours of spoken words, just as OpenAI's ChatGPT and other chatbots learn their skills by analyzing vast amounts of text retrieved from the Internet.
Luove admitted that he trained his technology using thousands of hours of recordings of thousands of voices, according to correspondence in the suit.
Mr Case, the lawyer representing Luvo, said the company trained its AI system using audio from a freely available database of English recordings called Openslr.org. He declined to comment when asked if the voice recordings of Mr. Lehrman and Ms. Sage were used to train the technology.
“We hope to control our voices, who we are, our careers,” Mr. Lehmann said. “We want to represent others that this has happened to and that will happen to them if nothing changes.”