AI & Agentic Systems

Hospital Voice Agent

A trilingual hospital receptionist that books appointments from inbound calls.

Software build · AI Voice Reception · FastAPI + Sarvam STT/TTS + Groq + Google Calendar

Visit live website

Hospital Voice Agent case study visual

Project illustration. Not a verified production screenshot.

Engagement: Software build. The case study documents the implementation scope. Live deployment status and usage are not asserted.

The capabilities below describe the work delivered. Quantified business results have not been published.

Context

This project explores inbound appointment reception in Hindi, Odia, and English, with bookings recorded in Google Calendar.

Challenge

The call flow needs to understand spoken requests, check available times, and confirm a booking while keeping the speech and calendar tools coordinated.

Solution

Avlys built a FastAPI call server with Sarvam speech, Groq tool-calling, and Google Calendar booking, plus a text test path so the flow can be checked without a live trunk.

Inside the project

System architecture.

Logical components and integration boundaries. This diagram is not a deployment or compliance certification.

Hospital Voice Agent layered architecture diagram
Open full-size diagram ↗
Architecture explained in text

A caller reaches the FastAPI voice endpoint through an Exotel or Plivo telephony adapter. Sarvam transcribes the audio; Groq handles the conversation and invokes availability or booking tools against Google Calendar. Sarvam synthesizes the response, which returns through FastAPI and telephony. Hospital configuration supplies hours, doctors, and prompts; a text endpoint supports conversation checks.

Delivered.

Inbound call flow

Hindi / Odia / English handling

Google Calendar booking tools

Text-only test endpoint

Capabilities delivered.

After-hours calls can still request an appointment.

Bookings land on the hospital calendar instead of a voicemail.

The same agent can take Hindi, Odia, or English without a fork.

Staff can rehearse the flow in text before connecting telephony.

How it was built.

01

Defined configurable hospital hours, doctors, and conversation prompts.

02

Connected audio input and output through FastAPI and Sarvam speech services.

03

Added Groq tool calls for calendar availability and appointment creation.

04

Provided a text conversation endpoint for testing before telephony integration.

Let’s scope your next system.

30 minutes, engineers on the call, no deck. We’ll map your workflow and tell you honestly whether AI pays back - and what a fixed-price pilot would look like.

Tell us about the project

Request a call or share your requirements. We reply within one business day.