Recently, OpenAI, the makers of ChatGPT, introduced their new Voice Engine technology that can replicate a person’s voice. While the potential applications of this technology are vast, OpenAI has decided not to release it publicly due to safety concerns. The company claims that it can clone a person’s voice with just 15 seconds of recording of that person speaking.
Due to the dangers of potential misuse, OpenAI plans to preview the Voice Engine technology with early testers but will not widely release it at this time. The AI company is aware of the risks associated with generating speech that resembles people’s voices, especially during an election year. Recent reports have shown that AI-generated voices were used in robocalls sent to voters in New Hampshire just before the presidential primary, highlighting the need for caution when it comes to voice-cloning technology.
Unlike other startup companies already selling similar voice-cloning technology, OpenAI’s approach with the Voice Engine involves early testers agreeing not to impersonate a person without their consent. Similarly, OpenAI has taken a cautious approach with its video-generator Sora, announcing but not widely releasing the technology. While OpenAI has not released the Voice Engine technology to the public yet, a trademark application filed in March suggests that the company is looking to get into the business of speech recognition and digital voice assistant technology. This move could potentially position OpenAI to compete with existing voice products like Amazon’s Alexa in the future.
