Add Telephony to a Gemini Live Agent
Skills:
Tool Use & Function Calling80%
Thor from Google DeepMind walks through how to set up the Gemini Live API with Twilio and deploy it to Google Cloud Run, so anyone can call your Gemini voice agent from a regular phone number. Inbound and outbound calls, real-time audio streaming, and the same agent running in the browser and on the phone from a single deployment.
What's covered: Using the GenAI Python SDK Twilio example from GitHub, building inbound and outbound call handlers with FastAPI and WebSockets, handling audio format conversion between Gemini and Twilio, deploying to Google Cloud Run with Secret Manager, and using Antigravity coding agent skills to accelerate the build.
]
Grab your Gemini API key at Google AI Studio and your Twilio credentials at twilio.com to get started. Full example and setup instructions linked below.
Resources:
Find the tutorial →https://goo.gle/4cokC0t
Find the code→https://goo.gle/4tYFYXV
What are you building with Gemini Live? Drop it in the comments.
Chapters
0:00 - Live Demo: Calling the Gemini Assistant on the Phone
0:51 - Introduction to Gemini 3.1 Flash Live & Twilio
1:16 - GitHub Examples & The Gen AI Python SDK
1:46 - Building the Integration with Coding Agents
2:57 - Code Walkthrough: FastAPI & WebSockets Setup
3:43 - Handling Twilio Inbound & Outbound Endpoints
5:25 - Audio Conversion (PCM to mu-law) & Partner Integrations
6:50 - Deployment to Google Cloud Run
7:35 - Web Browser Client Demo: Accents & Camera Vision
Speaker: Thor Schaeff
Products Mentioned: Google AI, Gemini
Watch on YouTube ↗
(saves to browser)
Sign in to unlock AI tutor explanation · ⚡30
More on: Tool Use & Function Calling
View skill →Related AI Lessons
⚡
⚡
⚡
⚡
From developer desks to the whole organization: Running Claude Cowork in Amazon Bedrock
AWS Machine Learning
MCP vs CLI for AI Agents: A Real AWS Benchmark (and Why the Popular Narrative Asks the Wrong Question)
Dev.to AI
AI research lab NeoCognition lands $40M seed to build agents that learn like humans
TechCrunch AI
LangGraph Error Handling Patterns for Production AI Agents
Dev.to AI
Chapters (9)
Live Demo: Calling the Gemini Assistant on the Phone
0:51
Introduction to Gemini 3.1 Flash Live & Twilio
1:16
GitHub Examples & The Gen AI Python SDK
1:46
Building the Integration with Coding Agents
2:57
Code Walkthrough: FastAPI & WebSockets Setup
3:43
Handling Twilio Inbound & Outbound Endpoints
5:25
Audio Conversion (PCM to mu-law) & Partner Integrations
6:50
Deployment to Google Cloud Run
7:35
Web Browser Client Demo: Accents & Camera Vision
🎓
Tutor Explanation
DeepCamp AI