OpenAI Conducts New Audio Cloning Tool Trials To Share Languages

Photo: OpenAI company illustration (photo: Unsplash)

JAKARTA - The company behind ChatGPT, OpenAI announced that it is conducting small-scale trials for its new artificial intelligence (AI) technology tool called Voice Engine.

"Voice Engine is a continuation of our commitment to understanding technical limitations and openly sharing what can be done with AI," OpenAI wrote in its official announcement quoted on Sunday, March 31.

Because it's still in a small-scale preview, Voice Engine can clone a 15-second audio into different languages from its original audio including Japan, Spain, Portuguese, Mandarin, Germany, and France with very natural speech.

We're sharing our learnings from a small-scale preview of Voice Engine, a model which uses text input and a single 15-second audio sample to generate natural-sounding speech that closely resembles the original speaker. https://t.co/yLsfGaVtrZ

— OpenAI (@OpenAI) March 29, 2024

We're sharing our learnings from a small-scale preview of Voice Engine, a model which uses text input and a single 15-second audio sample to generate natural-sounding speech that closely resembles the original speaker. https://t.co/yLsfGaVtrZ

The Voice Engine tool was first developed at the end of 2022, and has used it to support preset sounds available on the API text-to-speech and ChatGPT Voice and Read Aloud.

Despite the huge potential for synthetic or AI-generated noise abuse, OpenAI hopes that this tool can be used by the community responsibly and adapt to this new capability.

Understanding the serious risks that will arise, OpenAI claims to have collaborated with US and international partners from various levels of government, media, entertainment, education, civil society, and others to ensure this tool is safely built.

"Mitra, which tested Voice Engine today, has approved our use policy, which prohibits imitating other individuals or organizations without approval or legal rights," the company added.

Lastly, OpenAI million has implemented a series of security measures, including providing watermarks to track the origin of any audio generated by Voice Engine, as well as proactive monitoring of its use.

However, OpenAI did not mention when this new tool would be widely released to users. So, let's wait for the next information in the future.