ElevenLabs Launches 11ai: A Voice Assistant Built to Act on Your Workflow
Translate this article
ElevenLabs has released 11ai (alpha), a voice-first assistant designed to demonstrate the capabilities of its conversational AI platform. Unlike many current voice tools, which primarily respond to queries, 11ai is intended to interact with external services and initiate simple actions through user-authorized integrations.
The assistant is powered by the Model Context Protocol (MCP), a standardized approach for connecting AI agents to external APIs. This architecture allows 11ai to perform tasks across tools like Slack, Linear, Perplexity, and Notion, depending on configuration and permissions.
Voice Interaction with Task Execution in Mind
While most voice assistants excel at handling general knowledge queries or basic commands, they often fall short when it comes to meaningful task execution. 11ai attempts to close this gap by coordinating voice input with tool-specific actions — such as:
Integration via MCP
11ai’s support for MCP means it can interface with external systems through a unified protocol. The current alpha version includes default integrations with:
Available Now in Alpha
11ai is currently available as an alpha-stage proof of concept. ElevenLabs is offering free access during this period to gather user feedback and understand how voice-first interaction can support productivity.
To try 11ai:
1. Visit 11.ai to sign up
2. Choose a voice from the library or create a custom voice clone
3. Connect your tools, such as Google Calendar, Slack, or Perplexity
4. Initiate a voice session, starting with common workflows like task planning or summarizing information
ElevenLabs is particularly interested in hearing which integrations are most useful, what additional services should be supported, and how the voice interaction experience compares to traditional interfaces.
11ai serves as a technical demonstration of what voice interfaces could enable when combined with structured tool integration. It’s not a full-fledged assistant yet — but it offers a look into how natural language interaction might evolve to support everyday workflows, beyond answering questions.
About the Author
Chinedu Chimamora
Recent Articles
Subscribe to Newsletter
Enter your email address to register to our newsletter subscription!