Callsynthra AI Voice Calling Agent for Restaurants & Clinics | Twilio , RAG , LangGraph, OpenAI LLM

About this video

• Video Title: Callsynthra AI Voice Calling Agent for Restaurants & Clinics | Twilio , RAG , LangGraph, OpenAI LLM
• Channel: Habib Qureshi
• Speakers: Habib (implied from the transcript's first sentence)
• Duration: 00:07:10

Overview

This video demonstrates Callsynthra, an AI voice calling agent designed for businesses like restaurants and clinics. The presenter, Habib, walks through the product's features, including its ability to take orders, book reservations, manage cancellations, and answer customer queries using a knowledge base. The demonstration includes live calls to showcase the agent's functionality.

Key takeaways

• Callsynthra Functionality: The AI agent can handle various tasks for businesses, such as taking food orders, booking reservations, and processing cancellations.
• Dashboard and Metrics: The system provides a dashboard with key metrics like total calls, conversation rates, and a record of recent calls with transcriptions, recordings, and AI summaries.
• Knowledge Base Integration: Businesses can upload documents (like PDFs) to a knowledge base, allowing the AI to answer customer questions specific to the business.
• Use Case Scenarios: The video demonstrates placing an order, canceling an order (and the system's logic for when cancellation is not possible), and booking a table reservation.
• Customization: The AI agent can be configured and trained on specific business data to provide relevant responses.

Here is the transcript of the video:

Hi, my name is Habib and today I'm going to walk you through of our product called Centra which is an voice agent that we have built inhouse. It can work for multiple businesses like uh restaurants, clinics and much more. Uh today I'm going to walk you through that. So let's start. Okay. Uh so uh this is uh called Santra voice calling agent. Uh right now it is configured for restaurants and I'll walk you through the dashboard. So this is a couple of metrics. It it holds like total number of calls, every iteration, conversation rate and it also show you recent calls. There's a specific tab uh in which you can see each and every call what is happening the call transcription call recording AI summary. So right now this is for this is configured for restaurant and uh it is it can take order on the behalf of user and it can also book a reservation. So uh also this have a knowledge based system in which you can uh configure couple of things like a greeting message. So whenever user call on your business number it going to welcome with this message and it has knowledge base as well in which you can uh upload your um documents PDFs and everything and a going to answer according to your business. Okay. So uh let's call on this business and test the call.

Hi. All right. So I'm going to call on this business number and let's order something.

Hi Habib. Welcome to Burger Hub restaurant. How can I help you today?

Hi. Uh do you have anything for uh three people?

Yes, we have a family meal that serves multiple people. It includes 10 pieces of chicken, fries, and a one liter drink for $17.99. Would you like to order that?

Yes, please add this into the bucket.

Let me create your order real quick. I've added one X family meal to your order. The current total is $17.99. Would you like to proceed with the order?

Uh, actually, I just want one uh Zinger burger as well.

One moment. I'm updating your order. I've added one XZinger burger to your order. The current total is $23.98. Would you like to proceed with the order?

Yes.

Almost done confirming everything now. Your order has been confirmed. The total is $23.98 and your order number is number 25. Thank you for your order, Habib.

Thank you so much.

All right. So, yeah, this is uh this is how we can place an order. Okay. So, our order is placed and let's go into the order tab and see. Okay. So you can see uh this is the order that uh I have placed and if you can see I got one family meal, one single burger and this is my total and this is going to uh it's my delivery address. So actually the delivery address was already added. So if the customer is calling for the first time, it going to take the name and the address. Okay. And if you go to the calls, you can see the transcription. uh it has the recording as well and as well as AI summary in which saying total number of messages exchange average which response and uh the percentage of agent and customer talking. Okay. So we can also uh call cancel this order and uh let's try that.

Hihabib welcome to Burger Hub restaurant. How can I help you today? Hi. So, I just placed my order. My order number is 25 and I I don't want it. Uh I want to cancel it.

One moment. I'll handle the cancellation. Just to confirm, would you like to cancel your order number 25?

Yes.

Updating your order. Your order number 25 has been successfully cancelled. If you need any further assistance, feel free to ask. Thank you.

So, uh as you can see, uh the order is canceled. Okay. So, uh now the order is in preparation state and obviously we cannot cancel it. That's a really use case. And now let's try to uh call the session and try to cancel this order.

Hi Khabib, welcome to Burger Hub restaurant. How can I help you today? Hi, uh I want to cancel my order. My order number is 25.

One moment, I'll handle the cancellation. Just to confirm, would you like to proceed with cancelling your order number 25?

Yes.

One moment, I'll handle the cancellation. It seems that order number 25 is not in a confirmed state. If you need assistance with anything else or have another order you'd like to check, please let me know.

All right. Thank you. Okay. So you can see uh if the auditor is in the probation state, we cannot cancel it. Okay. So we have tried multiple things like placing an order, canceling an order and canceling an order which is in the uh the preparation state. And now let's try uh book an appointment because in this uh in this business you can also reserve a table. So let's call and reserve a table.

Hi, Habib. Welcome to Burger Hub restaurant. How can I help you today?

Hi, I want to book a table for four people. Tomorrow, 5 p.m.

One moment while I check availability. Your table for four people has been successfully reserved for tomorrow, April 30th, 2026 at 5:00 p.m. on the rooftop table T8. Enjoy your time.

All right. Thank you.

You're welcome.

Okay. So, you can see uh my resolution have been booked for four people. Table 09. Okay. So, yeah, this is uh this is a little bit about the software. Uh it can be configured with uh different residents. You can also train uh it can it can be trained on your data. So, so customer can ask questions according according to your business. For example, uh what are your opening hours? What do you serve? Uh do you have uh uh do you have outdoor setting and things like that? So you can train it and you can put all the information in your knowledge base and your AI going to answer according to this. All right. Thank you. Uh uh this is a little introduction and walk through of this software. Thank you.