Skip to main content

Just Say It: Build Your Shopping List with Your Voice and AI 🎀

Β· ChibiCart Team Β· 5 min read
Chibi girl speaking to her phone in a cozy kitchen β€” kawaii grocery items and AI robot appear as she builds her shopping list by voice

You're in the middle of cooking dinner. Both hands are covered in flour. You suddenly realize you're out of eggs, butter, and that one spice you always forget. What do you do?

In the old world: you'd wipe your hands, unlock your phone, open the app, and type each item one by one. By the time you're done, the onions are burning.

In the ChibiCart world: you tap the mic, say "I need eggs, butter, cumin, and olive oil" β€” and all four items are on your list before you can reach for the spatula. πŸ₯„

How It Works: Speak Naturally, Let AI Do the Rest 🧠

Most voice features make you speak in rigid commands. "Add. Milk. To. List." Robotic. Awkward. Nobody talks like that.

ChibiCart's voice input is different. You can speak the way you actually think:

πŸ—£οΈ "Today I wanna buy milk, beef, ground pork, and salad for my party"

β†’ ChibiCart adds: milk, beef, ground pork, salad
(Strips "today", "I wanna buy", "for my party" β€” keeps only what matters)

πŸ—£οΈ "Can you get 2 liters of milk, a dozen eggs, and some sourdough bread?"

β†’ ChibiCart adds: milk (2 liters), eggs (12), sourdough bread
(Quantities and units are extracted automatically)

The AI understands context, quantities, units, and filler words. You don't have to think about format β€” just talk.

The Tech Behind the Magic βš™οΈ

We built this with two layers working together β€” and deliberately chose tools that keep it fast, private, and free for all users.

πŸŽ™οΈ Layer 1: Web Speech API (Speech-to-Text)

Your browser's built-in speech recognition converts your voice to text in real-time β€” completely on-device, no audio ever sent to a server. You see your words appear live as you speak. Works on Chrome, Safari (including iOS), and Edge.

πŸ€– Layer 2: Gemini AI (Natural Language Understanding)

Once you stop speaking, the transcript is sent to Google's Gemini Flash model β€” the same AI that powers ChibiCart's receipt scanning and image generation. It strips filler words, extracts item names, quantities, and units, and returns a clean structured list. The whole process takes about a second.

We also built in a smart pre-check: if you accidentally tap the mic and say something unrelated to shopping, Gemini recognizes it and gently lets you know β€” rather than adding "my weekend plans" to your grocery list. πŸ˜„

Real Scenarios Where This Saves You πŸ›’

🍳 Cooking and realizing you're out of things

Hands busy? Just say what you need. No typing, no unlocking, no interrupting your flow.

πŸš— Driving home and remembering errands

Pull over safely, tap the mic, rattle off everything you need. Done before the light turns green.

πŸ›οΈ Lying in bed thinking about tomorrow's grocery run

Don't get up. Just speak your list into the dark. It'll be there in the morning.

πŸ‘¨β€πŸ‘©β€πŸ‘§ Planning a big family meal

Rattle off 10 ingredients in one breath. Voice input handles the whole list at once β€” no tapping 10 times.

Speak Your Language 🌏

ChibiCart supports English, Simplified Chinese, and Traditional Chinese throughout the app β€” and voice input follows your language setting automatically. Switch the app to Chinese, and the microphone listens for Chinese speech. No configuration needed.

πŸ—£οΈ "ζˆ‘ιœ€θ¦δΉ°η‰›ε₯Άγ€ιΈ‘θ›‹ε’Œι’εŒ…"
β†’ Adds: milk, eggs, bread β€” recognized in Chinese, added to your list

Your Voice Stays Private πŸ”’

We know voice features can feel invasive. Here's exactly what happens with your audio:

  • Your audio is never recorded or stored β€” speech-to-text happens entirely in your browser using the Web Speech API
  • Only the text transcript is sent to Gemini for parsing β€” not audio
  • Microphone permission is requested by your browser, not ChibiCart β€” you can revoke it any time in browser settings
  • No account required to use voice input β€” it works for all users

How to Use It πŸ‘‡

  1. Open any shopping list in ChibiCart
  2. Tap the 🎀 microphone icon next to the input field
  3. Allow microphone access when your browser asks (first time only)
  4. Speak naturally β€” say everything you need in one go
  5. Stop speaking (or tap the mic again to stop early)
  6. Watch as AI parses your words and adds all items instantly

That's it. No special syntax. No commands to memorize. Just talk like you're telling a friend what to pick up.

Shopping Lists, Finally Hands-Free πŸ™Œ

Voice input is one of those features that sounds like a small convenience β€” until you use it while your hands are full, or you're rushing out the door, or you're trying to capture a whole week's meal plan in 30 seconds.

Then it becomes indispensable.

We built it to be fast, private, multilingual, and smart enough to understand how people actually speak β€” not how apps want them to speak. Because the best technology gets out of your way and just works.

Give it a try. Your grocery list is one sentence away. πŸ›’βœ¨

Try Voice Input Now 🎀

Open a shopping list and tap the microphone. Say what you need β€” AI handles the rest.

Open ChibiCart

Written by the ChibiCart Team
Making shopping effortless, one feature at a time πŸ›’πŸŽ€βœ¨