OpenAI, the company behind ChatGPT, is foraying into the voice assistant industry with its new technology that can replicate a person’s voice. The Voice Engine technology can reportedly clone a voice using just a 15-second recording of the individual speaking. However, despite this advancement, OpenAI has decided not to release the technology publicly at this time due to concerns about potential misuse.
The company recently filed a trademark application for Voice Engine and plans to share it with early testers before a wider release. They are taking a cautious approach, acknowledging the risks associated with generating speech that closely resembles real voices, especially during elections like the upcoming presidential race.
In light of recent incidents involving AI-generated robocalls mimicking public figures, such as President Joe Biden, OpenAI is conscious of the potential for misuse and is taking steps to ensure responsible usage. While various companies offer voice-cloning technology, OpenAI’s focus on safety and ethical considerations sets them apart.
Early testers of Voice Engine have agreed not to impersonate others without their permission and must disclose that the voices generated are artificial. This responsible stance reflects OpenAI’s commitment to ethical AI practices, as seen in their previous projects like the chatbot and image generator DALL-E.
With a trademark application hinting at broader speech recognition and digital voice assistant ambitions, OpenAI is positioning itself to compete with established products like Amazon’s Alexa in the future. Keep an eye on the latest developments in AI by subscribing to the Eye on AI newsletter for free.
