How to Orchestrate Multi-Call Conversations with an LLM and Twilio Conversation Memory with PHP

Summary

This tutorial shows how to build a PHP voice agent that retains caller context, preferences, and action history across separate inbound calls. It uses Twilio Conversation Memory with a Memory Store, an Open Swoole service, Twilio ConversationRelay, and an OpenAI integration using gpt-4o-mini; the example models an auto-repair shop assistant. The OpenAI service appends supplied prior customer context to its system prompt, while streamed responses are sent through a WebSocket and environment variables provide API credentials and the Memory Store ID. After running the service on port 9501, exposing it with ngrok, and configuring the phone-number webhook, the tutorial tests two calls from the same number and says the second call should remember details from the first.

Takeaways

1. 1

   Twilio Conversation Memory persists context between separate calls, allowing a returning caller’s previous discussion to be used without repeating details.

2. 2

   The OpenAI integration uses gpt-4o-mini with a 400-token limit and streams generated content through ConversationRelayHandler over a WebSocket.

3. 3

   Conversation Memory is not intended for sensitive information, and testing requires two calls from the same phone number after configuring the Twilio webhook.

FAQ