OpenAI Decides Against Releasing Voice Engine Tool
Deeming the tool that can generate a convincing clone of anyone’s voice using just 15 seconds of recorded audio too risky for general release, OpenAI said it was seeking to minimise the threat of damaging misinformation ahead of election year.
The ‘Dangerous’ Voice Clone Tool
As per The Guardian, Voice Engine was first developed in 2022. An initial version of the tool was used for OpenAI’s AI tool ChatGPT’s text-to-speech feature. However, after realising how powerful the tool is, keeping with its “cautious and informed” approach, the company has revealed its power publicly as it seeks to release the tool more widely. “We hope to start a dialogue on the responsible deployment of synthetic voices, and how society can adapt to these new capabilities,” said OpenAI in a blogpost. “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 added.
Voice Engine’s Uses
In its blog post, OpenAI gave examples of real-world uses of the technology. This includes generating scripted voiceovers and preserving accents in translations. OpenAI said it was opting to “preview but not widely release this technology at this time,” in order “to bolster societal resilience against the challenges brought by ever more convincing generative models.” It added that it is encouraging “steps like phasing out voice-based authentication as a security measure for accessing bank accounts and other sensitive information.” The company also called for the exploration of “policies to protect the use of individuals’ voices in AI,” and “educating the public in understanding the capabilities and limitations of AI technologies, including the possibility of deceptive AI content.” OpenAI said that the generated audio will be watermarked so that its origins can be traced. “Our terms with these (current) partners require explicit and informed consent from the original speaker, and we don’t allow developers to build ways for individual users to create their own voices.”
Competitors in the Field
While OpenAI’s Voice Engine stands out for its technical simplicity and efficiency, competitors offering similar tools are already available to the public. Companies like ElevenLabs can generate complete voice clones with just a few minutes of audio.
