Try the New 2X Voices Using a Speech-to-Speech Model

We’re excited to announce that customers can now preview the next generation of the 2X speech-to-speech voices — designed to deliver more natural, expressive, and human-like conversations than ever b…

Updated 6 hours ago by Kevin DeMeritt

We’re excited to announce that customers can now preview the next generation of the 2X speech-to-speech voices — designed to deliver more natural, expressive, and human-like conversations than ever before.

With just a few clicks, you can experience live voice demos and explore how our updated models handle tone, rhythm, and emotion in real time. These voices offer a significant leap forward in realism, responsiveness, and adaptability.

Smarter Conversations That Go the Distance

The new voices aren’t just clearer — they’re smarter and more capable. Our latest models can sustain longer, more natural dialogue without losing context or emotional consistency. They’re able to detect conversational cues that signal hesitation or disinterest, helping your AI respond more empathetically and guide the discussion toward re-engagement — a key advantage in overcoming objections or uncertainty.

Faster, Clearer, and More Resilient

We’ve also enhanced real-time performance so conversations flow more smoothly than ever. The voices now respond faster, creating a more natural back-and-forth that feels immediate and intuitive. Improved background noise handling ensures clarity even in dynamic or noisy environments, making the experience more reliable and lifelike across all use cases.