Skip to content

NEjim2001/bud-dial

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

💻 Devpost

BudDial on Devpost

BudDial

BudDial is a mobile-first, AI-powered voice assistant that automates real-world appointment booking for users via structured calls, turning stressful phone calls into seamless one-click interactions.

Overview

BudDial allows users to request appointments or reservations through structured forms. Once submitted, an AI-powered assistant makes a real phone call to the business, confirms details, and returns results to the user.

  • Guided Request: Users select a business on the map and complete a short form (e.g., time, service, party size).
  • AI Execution: The AI Buddy calls the business through Twilio and speaks naturally using ElevenLabs voice synthesis.
  • Result Handling: When the call ends, Gemini processes the transcript to extract structured booking data.
  • Feedback: Users receive either a Success screen with confirmation details or a Failure screen with clear, actionable reasons.

Technologies

Mobile Frontend

  • Framework: React Native
  • Design: Figma prototypes focused on accessibility, UX/UI experience

Backend & AI Orchestration

  • Server: Node.js with Express
  • Telephony: Twilio API for initiating and managing calls
  • Voice: ElevenLabs for natural Text-to-Speech
  • LLM Processing: Gemini for transcript parsing and structured output
  • Maps Integration: Google Maps API for verified business data (names, numbers, addresses)

Technical Workflow

  1. The mobile app sends a POST /api/call request to initiate an AI-assisted call.
  2. The backend uses Twilio and ElevenLabs to execute the voice conversation.
  3. When the call completes, the transcript is sent to Gemini for analysis.
  4. Gemini returns a structured JSON response with extracted booking data.
  5. The result is displayed to the user as either a success confirmation or a failure report.

Achievements

  • Built a full AI voice pipeline requiring no human intervention from request to completion.
  • Achieved consistent, structured output from raw phone call transcripts.
  • Designed a reliable AI voice persona that improved user trust and booking success rates.

Future Improvements

  • Calendar Integration: Sync successful bookings directly with user calendars.
  • Async Booking Mode: Allow users to submit requests that the AI executes later.

About

HackUTA project

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •