From Decrypt Media: 2024-03-29 21:57:03

OpenAI introduces Voice Engine, a voice cloning technology that can replicate human speech patterns with a small audio sample. ElevenLabs, in comparison, requires at least a one-minute sample for voice cloning. The technology can recreate voices, like that of a patient with impaired speech, making a huge impact.

Voice Engine enables text-to-speech with the patient’s own voice, showcasing its potential. OpenAI collaborates with Lifespan to bring this technology to life, emphasizing responsible deployment to prevent misuse. With safeguards in place and restrictions on emulation, OpenAI is paving the way for safe and ethical use of synthetic voices.

While OpenAI works on GPT-5 and the advanced video generator Sora, Voice Engine stands out among voice cloning tools. Outperforming other models, Voice Engine’s future deployment hinges on responsible use and collaboration with key partners. The company’s Q* project remains a mystery, focused on enhancing AI reasoning capabilities.



Read more at Decrypt Media: New ‘Voice Engine’ from OpenAI Needs Only 15 Seconds to Clone Speech