xAI Launches Voice Agent Builder
Key point
xAI has released a beta version of Voice Agent Builder, which lets users build high-performance voice agents in 2 minutes without writing a single line of code.
Details
xAI has unveiled Voice Agent Builder, a no-code platform built on Grok Voice, in beta. This platform brings together phone connectivity, knowledge retrieval, tool use, guardrails, and observability all in one place without the need to build a complex stack, helping operators and developers build production-level voice agents instantly.
Unlike existing voice stacks that connect STT, LLM, and TTS from separate providers—causing latency and cost issues—Voice Agent Builder maximizes efficiency by using a Speech-to-Speech pipeline tightly coupled with Grok Voice.
Key features include the following:
- Knowledge Base: Upload various document formats so the agent can learn and retrieve business information.
- Tools & Connectors: Perform real actions such as bookings, refund processing, and information lookup through Google Calendar, Outlook, APIs, web search, and X search.
- Custom Voices: Use more than 80 built-in voices, or clone a brand's own unique voice with about 2 minutes of audio.
- Flexible Connectivity: A free phone number is provided, and the platform also supports SIP connections for existing numbers and client connections via WebSocket.
On τ-voice Bench, which measures noise and interruption handling in real call environments as well as multilingual capability, Grok Voice Think Fast 1.0 scored 67.3%, significantly outperforming Gemini 3.1 Flash Live (43.8%) and GPT Realtime 1.5 (35.3%).
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.