ElevenLabs has unveiled a new conversational AI platform today that lets you create custom voice agents for applications ranging from customer service to interactive gaming characters, marking a significant push into the growing AI voice technology market.
Key Points
- The platform supports 31 languages and offers integration with major AI models like Gemini, Claude, and GPT
- Companies can deploy agents at $0.10 per minute on business plans, with rates dropping to $0.015 per minute at scale
- The technology includes real-time interruption detection and turn-taking features to make conversations feel more natural
The platform combines speech recognition, text generation, and voice synthesis with advanced features like interruption handling and turn-taking capabilities. This means your AI agent can pause when interrupted and respond naturally in conversations, similar to human interaction patterns.
"We've simplified the technical setup so you can focus on customizing your agent," said Jozef, ElevenLabs' lead engineer. "Our platform eliminates months of development time typically spent building conversation stacks from scratch."
The technology supports 31 languages and integrates with major AI models including Google's Gemini, Anthropic's Claude, and OpenAI's GPT. You can also connect your own custom AI implementation, giving businesses flexibility in how they deploy the technology.
Pricing starts at $0.10 per minute for business plans, with costs dropping to $0.015 per minute at scale. During the beta period, ElevenLabs is covering the additional costs of language model usage, though these will eventually be passed through to customers.
A key feature of the platform is its ability to handle real-world applications through native integration with Twilio for phone calls. This means businesses can deploy AI agents directly into their existing phone systems for customer service or outbound sales.
The launch comes as more companies seek to automate customer interactions while maintaining natural-feeling conversations. ElevenLabs' platform includes monitoring tools that provide full transcripts and automated evaluation of conversations, helping businesses ensure quality as they scale their AI deployments.
For developers, the platform offers SDKs for Python, JavaScript, React, and Swift, along with direct WebSocket API access. Companies can also import their existing documentation to build the AI agent's knowledge base about their specific products and services.