TaskVoice AI

AI-powered task manager with voice and chat input.

Comment

Inspiration

Managing daily tasks can feel like a chore especially when switching between tools or typing things out manually. As a freelancer, I often juggle multiple clients and projects, and I’ve personally struggled with time management, scheduling, sending updates on time, and just keeping track of what needs to be done.

I wanted to build something that lets people offload that mental load using simple voice or chat interaction. That's where TaskVoice AI was born a smart productivity assistant that lets you create tasks using AI chat, AI voice, or manual input, however you prefer to work.

I was inspired by the way people talk to assistants like Siri or ChatGPT, but I wanted to give that power a focused purpose: getting things done fast and frictionlessly.

What it does

TaskVoice AI is a smart productivity assistant that helps users create and manage tasks using three input modes:

Voice AI — Speak your tasks naturally, and it understands and schedules them. Chat AI — Type your task in plain English; the AI handles the rest. Manual input — Quickly add or customize tasks with full control.

Tasks are automatically categorized into Today, Tomorrow, or Upcoming based on natural language. It's fast, flexible, and works the way you think — by voice, chat, or tap.

How we built it

This application is built using a modern web development stack, combining powerful tools for AI, voice interaction, and real-time task management.

🧱 Tech Stack Overview

  • Languages: TypeScript
  • Frontend Framework: React
  • Styling: Tailwind CSS
  • Routing: React Router DOM
  • Icons: Lucide React
  • Date Manipulation: date-fns
  • Unique IDs: uuid
  • Build Tool: Vite

☁️ Backend & Infrastructure

  • Cloud Database: Supabase (PostgreSQL)
  • Authentication: Supabase Auth with Row Level Security (RLS)

🧠 AI & Voice Integration

  • Conversational AI: OpenAI API
  • Text-to-Speech (TTS): ElevenLabs API
  • Speech-to-Text (STT): Google Speech API

We used Bolt’s powerful prompt system to structure and fine-tune AI responses. This made it easy to extract intent, detect time expressions (like “tomorrow” or “after lunch”), and trigger smart task actions in real-time.

🧼 Code Quality & Deployment

  • Linting: ESLint
  • Deployment: Netlify

All components were built with a modular, scalable mindset ensuring smooth UX and fast performance across voice, chat, and manual task flows. Leveraging Bolt allowed us to iterate quickly while keeping AI behavior predictable and aligned with the user experience.

Challenges we ran into

Building TaskVoice AI in a few days came with several technical and UX hurdles: Voice latency for first response: The voice AI had a 2–4 second delay when replying for the first time. Solving this while keeping the experience smooth was tricky. Parsing natural time expressions: Converting vague phrases like “after lunch” or “next Friday morning” into accurate timestamps required tuning prompts and logic carefully. Syncing voice, chat, and manual flows: Ensuring that all 3 input modes worked independently yet shared task logic took thoughtful state management. Real-time voice transcription display: Showing live speech-to-text results while also processing them in the backend had to be snappy and accurate to feel natural. Prompt testing and tuning: I spent a lot of time experimenting with Bolt prompt blocks to extract just the right information from voice/chat inputs and route them into the task list cleanly. Hackathon time crunch: Balancing polish, functionality, and ambitious ideas like AI voice required rapid iteration and ruthless prioritization.

Accomplishments that we're proud of

  • Built a voice assistant that adds tasks in real time from natural speech
  • AI chat understands typed input like “Call John tomorrow at 2 PM”
  • Tasks are auto-sorted into Today, Tomorrow, and Upcoming
  • Integrated OpenAI, ElevenLabs, and Google Speech APIs
  • Used Bolt's prompt system for smart task extraction
  • Polished UI with React, Tailwind, and smooth interactions
  • Completed the full MVP in under 5 days

What we learned

  • How to connect and fine-tune OpenAI prompts for real-world task creation
  • Real-world use of voice input using ElevenLabs + Google Speech API
  • Managing real-time state between inputs and UI dynamically
  • UI design techniques to keep interactions clean even with 3 input modes
  • The power of making AI interactions feel invisible and helpful not just smart

What's next for TaskVoice AI

Reminders & notifications – Get notified before important deadlines Mobile-first experience – With PWA support for offline use AI-generated task suggestions – Based on habits and patterns Chat history & voice command logs – So users can track previous interactions Google Calendar Integration – Auto-sync tasks and meetings directly to your calendar Client Communication Tools – Automate sending task updates to freelancer clients via email Meeting Automation – Use AI to suggest and schedule meetings based on your availability and deadlines Smart Notes & Context – Let users attach voice or text notes to tasks for better clarity The vision is to create a lightweight, voice-first productivity system that integrates seamlessly into daily life especially for freelancers, remote workers, and busy professionals.

Built With

Share this project:

Updates