OpenAI is giving developers access to its advanced speech AI engine, which uses the same technology as ChatGPT’s voice mode. This move is expected to lead to a surge in AI applications featuring conversational voice interfaces, enhancing users' interactions with technology.
The new speech feature was announced alongside other tools at OpenAI’s DevDay event in San Francisco. Early adopters include the fitness and nutrition app Healthify and Speak, a language learning app. Another capability introduced at the event allows developers to fine-tune AI models using pictures, expanding the potential for creative and personalized applications.
During a demo, OpenAI showed how its new audio tools could be combined with other services like Twilio’s API. An AI assistant called a fictional candy shop to order 400 chocolate-covered strawberries in the example. Only the voices offered by OpenAI, the same ones used in ChatGPT, will be available for developers. Although these voices won’t have watermarks, OpenAI's terms of service prohibit using its technology to mislead or spam others.
These announcements are pivotal for OpenAI, which is making headlines for its massive fundraising efforts. It also follows the recent departure of CTO Mira Murati and two other top executives, highlighting a significant change for the company.