Announcing the launch of next-generation audio models via the API
A new generation of audio models is now available via API, giving developers unprecedented control over text-to-speech output. For the first time, they can direct how a TTS model delivers its lines using natural language instructions—such as asking it to sound like a caring customer support representative—enabling highly customized voice-driven agents and applications.
For the first time, developers are able to guide how a text-to-speech model delivers its lines—say, by asking it to "sound like a caring customer support representative"—bringing a whole new degree of customization to voice-driven agents.
Comments (0)