OpenAI has unveiled an AI tool, Voice Engine, that can emotionally and realistically recite texts in anyone’s voice based on a 15-second speech sample.
.in_text_content_22 { width: 300px; height: 600px; } @media(min-width: 600px) { .in_text_content_22 { width: 580px; height: 400px; } }
The tool has been created for about two years, but is not yet publicly available because it can be used to create deepfakes.
The Voice Engine service has been in development since late 2022—around the same time ChatGPT launched—and uses text input and a 15-second audio sample to generate “natural-sounding language that closely resembles the original speaker.”
It aims to help with reading, translation, support for non-speaking people and help patients with degenerative speech conditions regain their voice, and is also a service that helps speed up the delivery of public services in remote areas, it said.
Voice Engine connects to OpenAI’s suite of generative AI platforms, including Dall-E for images, Sora for video, and ChatGPT, kickstarting the generative AI revolution.
Early testers will be able to preview the Voice Engine, but the tool “has no plans to make it public at this time,” OpenAI said.
The company has not provided a general release schedule for Voice Engine.
Based on these conversations and the results of these small-scale tests, we will make a more informed decision about whether and how to deploy this technology at scale, it said.
Voice cloning has long been used in industries and supports tasks such as customer support and acquisition, as well as reducing costs and operational efficiency.
The global voice cloning market is projected to reach nearly $9.3 billion, according to the latest data from Grand View Research. by 2030 from approximately 1.45 billion in 2022, with an annual growth rate exceeding 26%.
However, with the advent of generative artificial intelligence, risks have also come to the fore, including disinformation, counterfeiting and bias, among others.
OpenAI in particular acknowledged that these risks are heightened in 2024 as it is an election year in the US, where a rematch is expected between incumbent President Joe Biden and Donald Trump, engaged in a highly polarizing fight fueled by social media.
We recognize that creating a broadcast that resonates with the people’s voices has significant risks…we work with U.S. and international partners in government, media, entertainment, education, civil society, and more to ensure we incorporate their feedback into our creation,” OpenAI said.
OpenAI said its Voice Engine partners have agreed to its usage policy, which prohibits impersonation of another person or entity without consent or legal right and requires “explicit and informed consent” from the original speaker.
We don’t allow developers to create ways for individual users to create their own voices. Partners must also clearly communicate to their audience that the voices they hear are created by artificial intelligence,” the company said.
Source: Racurs
I am David Wyatt, a professional writer and journalist for Buna Times. I specialize in the world section of news coverage, where I bring to light stories and issues that affect us globally. As a graduate of Journalism, I have always had the passion to spread knowledge through writing.