Introducing Conversational AI

Today we're proud to introduce Conversational AI - our all in one platform for building customizable, interactive voice agents. With Conversational AI, you can create outbound sales dialers, scheduling agents, interactive game characters, tutors, customer support agents, and more.

Our platform is designed for low latency, thanks to fewer server calls. Conversational AI handles turn-taking and interruptions gracefully using a real-time model to predict when a speaker is finished. We’ve also simplified the technical setup so you can focus on customizing your agent.

Conversational AI connects our life-like Text to Speech, Speech to Text, an LLM of your choice, with our custom turn-taking and interruption logic that makes conversation feel natural. This leaves you to focus on:

• Building your knowledge base
• Crafting your system prompt
• Connecting your apps with function calling
• Selecting your favorite voice from our library or cloning your own
• Finding ways to use Conversational AI to delight your customers with unbelievable experiences

Conversational AI includes advanced features to help you build more interactive agents, including:

• Native Twilio integration for handling calls
• Server-side and client-side tool calling for added flexibility
• Dynamic prompting to create personalized conversations

Here’s our overview on how to configure your first agent. If you prefer complete control, you can connect your own LLM or server with this guide. We have a Python, Javascript, React and Swift SDK, or you can integrate with the WebSocket API directly. 

You can see an example project built using our SDK here.

Want to know more about how we built Conversational AI? Hear from Jozef, our lead engineer, here.

Thinking of building your startup on Conversational AI? You may qualify for our free startup grant program. Learn more here.