Voice synthesis has come a great distance since 1978’s Speak & Spell toy, which as soon as wowed individuals with its state-of-the-art potential to learn phrases aloud utilizing an digital voice. Now, utilizing deep-learning AI fashions, software program can create not solely realistic-sounding voices however may convincingly imitate present voices utilizing small samples of audio.
Along these traces, OpenAI this week introduced Voice Engine, a text-to-speech AI mannequin for creating artificial voices primarily based on a 15-second section of recorded audio. It has offered audio samples of the Voice Engine in motion on its web site.
Once a voice is cloned, a consumer can enter textual content into the Voice Engine and get an AI-generated voice end result. But OpenAI is just not able to extensively launch its know-how. The firm initially deliberate to launch a pilot program for builders to enroll in the Voice Engine API earlier this month. But after extra consideration about moral implications, the corporate determined to cut back its ambitions for now.
“In line with our approach to AI safety and our voluntary commitments, we are choosing to preview but not widely release this technology at this time,” the corporate writes. “We hope this preview of Voice Engine both underscores its potential and also motivates the need to bolster societal resilience against the challenges brought by ever more convincing generative models.”
Voice cloning tech usually is just not notably new—there have been a number of AI voice synthesis fashions since 2022, and the tech is energetic within the open supply group with packages like OpenVoice and XTTSv2. But the concept that OpenAI is inching towards letting anybody use its explicit model of voice tech is notable. And in some methods, the corporate’s reticence to launch it absolutely is likely to be the larger story.
OpenAI says that advantages of its voice know-how embrace offering studying help by means of natural-sounding voices, enabling world attain for creators by translating content material whereas preserving native accents, supporting non-verbal people with customized speech choices, and helping sufferers in recovering their very own voice after speech-impairing situations.
But it additionally implies that anybody with 15 seconds of somebody’s recorded voice might successfully clone it, and that has apparent implications for potential misuse. Even if OpenAI by no means extensively releases its Voice Engine, the flexibility to clone voices has already brought on hassle in society by means of cellphone scams the place somebody imitates a beloved one’s voice and election marketing campaign robocalls that includes cloned voices from politicians like Joe Biden.
Also, researchers and reporters have proven that voice-cloning know-how can be utilized to interrupt into financial institution accounts that use voice authentication (reminiscent of Chase’s Voice ID), which prompted US senator Sherrod Brown of Ohio, the chair of the US Senate Committee on Banking, Housing, and Urban Affairs, to ship a letter to the CEOs of a number of main banks in May 2023 to inquire in regards to the safety measures banks are taking to counteract AI-powered dangers.
OpenAI acknowledges that the tech may trigger hassle if broadly launched, so it is initially making an attempt to work round these points with a algorithm. It has been testing the know-how with a set of choose associate corporations since final 12 months. For instance, video synthesis firm HeyGen has been utilizing the mannequin to translate a speaker’s voice into different languages whereas conserving the identical vocal sound.