Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Zero To Mastery

AI Voice Agents with AWS

via Zero To Mastery Path

Go to class Write review

Overview

Build real-time speech-to-speech agents from the ground up using Python and AWS Bedrock. This course walks you through the streaming architecture, audio debugging, and tool integrations that power modern voice AI.
  • Set up AWS Bedrock for real-time voice agent use
  • Stream bidirectional audio with low latency
  • Use Python asyncio to manage non-blocking tasks
  • Handle interruptible speech and voice interactions
  • Send and receive Bedrock streaming events
  • Build reusable audio and streaming manager classes
  • Integrate DynamoDB for real-time data updates
  • Test and run a complete, deploy-ready voice agent

Syllabus

  •   Introduction
    • Course Introduction
    • What We Are Building
    • Exercise: Meet Your Classmates and Instructor
    • Course Resources
    • Additional Course Information
  •   Setting up AWS
    • Setting Up AWS Access Keys
    • Setting Up Files
  •   Understanding Speech-to-Speech Models
    • Understanding Speech-to-Speech Models
    • Understanding Bidirectional Streaming
  •   Audio Configurations + Debugging Functions
    • Creating Audio Configurations
    • Setting Up Debugging Functions
  •   Python Event Loop for Async Tasks
    • Non-Blocking Asyncio Python
    • Eventloop and Multithreads in Python
  •   Voice Agent Tool Access
    • Getting Guests, Dynamodb Call
    • Getting Reservations, Dynamodb Call
    • Updating Reservations, Dynamodb Call
  •   Exploring Nova Sonic Templates
    • Event Templates Part 1
    • Event Templates Part 2
  •   Exploring Tool Calls
    • Exploring Tools Our Model Has Access To
    • Tool Result Event
  •   Initialising and Using the Bedrock Stream Manager Class
    • Initialising the Bedrock Stream Manager Class
    • Initialising the Bedrock Stream
    • Sending Raw Events to Bedrock
    • Processing Audio Input
    • Sending Events to the Bedrock Stream
    • Processing Incoming Responses From Bedrock
    • Handling Tool Requests + Completions
    • Executing Tools + Gracious Closing and Shutting Down
  •   Exploring the AudioStreamer Class
    • Separate Input and Output Streams
    • Finishing the Audio Streamer Class
    • Ending the Stream Clarification
    • Finishing Up Our Final Script
  •   Understanding AWS Quotas + Adding Credentials to our Script
    • AWS Quotas and Credentials
  •   Setting Up Our Final App
    • Installing Necessary Libraries
    • Setting up DynamoDB
    • First Test of Our Agent
    • Testing Reservation Updates
    • Testing with the Debug Flag
    • Testing the Final Product
  •   Cleanup and Congratulations
    • Cleaning Up
    • Congratulations!
  •   Where To Go From Here?
    • Review This Byte!

Taught by

Patrik Szepesi

Reviews

Start your review of AI Voice Agents with AWS

Go to class

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Sign up for free
Someone learning on their laptop while sitting on the floor.