[Full Workshop] Building Conversational AI Agents
summary
Hands-on walkthrough of a pipeline that transcribes speech, routes text to an LLM, and streams back synthetic speech with live language switching using 11 Labs tools: the ASR claims benchmark coverage across 99 languages with word-level timestamps and speaker diarization, a language-detection system tool that auto-switches voices, a voice library of over 5,000 voices (marketplace payouts > $5M), function-calling and RAG support for external tools, agent-to-agent transfers, safety watermarking, and pricing based on call minutes.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 0:26Welcome
- 2:25Resources
- 5:17Bark
- 10:59SpeechtoText
- 17:59Voice Library
- 20:40Demo
- 27:49Questions
- 29:14Agent Configuration
- 31:27Scheduling
- 33:39Best Practices
- 34:24Voice Models
- 35:21Pricing
- 37:41Agent to Agent Transfers
- 49:09Safety Tools
- 53:08Intermixed Languages
- 56:02Nvidia Tokyo
related talks
10mBuilding Conversational AI Agents at Scale
01h 58mWhat If Your Backend Could Talk? Building Real-Time AI Voice Agents
32mBuilding LLM Voice Bots with Open Source Tools
20m10 Things I Learned Building Voice Agents in 2025
08mGive Your Chat Agent a Voice
33mDesigning Voice Agents for Real Conversations
01h 04mBuilding Expertise Into Generated Conversations
12mReal-time voice agents with ElevenLabs and AI SDK