Loading post…
TwilioMatthew SetterTutorial
How to Augment Voice Calls with Twilio Intelligence
Summary
This tutorial shows PHP developers how to augment a Twilio voice AI agent with Conversation Intelligence, Conversation Orchestrator, and Conversation Memory. The goal is to retrieve a short summary of each customer call, analyze the caller’s sentiment, and check whether the agent followed set guidelines. The information is persisted to a SQLite database for later use. The tutorial builds on a previous voice agent tutorial and requires a Twilio account, a voice-capable phone number, and ngrok to expose the local server. It walks through setting up Conversation Orchestrator, Conversation Intelligence, and the application database, then testing the flow with example questions. The post also mentions the Conversations API (V2) for retrieving call summaries and sentiment analysis.
Context
The post is a tutorial for PHP developers extending a Twilio voice AI agent with Conversation Intelligence, Conversation Orchestrator, and Conversation Memory, persisting call summaries and sentiment to SQLite.
Approach / What changed
Set up Conversation Orchestrator and Conversation Memory, then Conversation Intelligence, make the app publicly accessible via ngrok, create a SQLite database with dump.sql, and test the flow with sample questions to verify summaries, sentiment, and guideline adherence.
Takeaways
Conversation Intelligence, Conversation Orchestrator, and Conversation Memory work together to provide real-time call summaries, sentiment analysis, and guideline adherence checks.
The Conversations API (V2) can retrieve a short summary of each call and an analysis of the caller’s sentiment, which can be persisted to a SQLite database.
The tutorial is geared toward PHP developers but offers C# and JavaScript options for building the same project.
FAQ
What do I need to follow this tutorial?
You need a free Twilio account, a voice-capable Twilio phone number, the Owl Air phone agent from the previous tutorial or any Conversation, SQLite support, and ngrok or a similar tool to expose your local server to Twilio.
What can I do with the retrieved conversation data?
You can retrieve a short summary of each customer call, an analysis of the caller's sentiment, and whether the agent followed the guidelines you set. This information can be persisted to a SQLite database for later use.
How do I test the application?
You can test the full flow by asking questions like 'What's the baggage policy?', 'How do loyalty points work?', and 'Can I change my flight?' to verify the agent's responses and the stored conversation data.
Source and freshness
- Source
- twilio.com
- Published
- Added to Yomu
Summaries are generated from the original by a language model. Check the original before relying on it.
Keep reading
Related reading
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
The best engineering reads, once a week
One useful email. Unsubscribe anytime.