Use Case
AI tools for restaurant websites
The MiOffice toolkit for cafes, bakeries, and restaurants — menu extraction, image pipeline, festival hero artwork, voice-search structured data. Used to build Pastry Corner Cafe & Lounge.
Who this is for
A restaurant website lives or dies on three things: how fast it loads on mobile, whether the menu shows up in search, and whether the imagery looks like the food. The seven tools below cover all three without leaving the browser — no desktop image editor, no separate OCR app, no licensed stock photo subscription.
Every tool runs client-side or dispatches to a dedicated GPU worker. Files stay private. There's no subscription — one credit pack covers a typical build.
The seven tools
- PDF to TextMenu extraction
Pull menu items, prices, and addon groups straight out of the restaurant’s existing PDF menu. Seeds the structured menu rows your site renders.
- AI Document SummarizerContent depth
Compress legacy catering brochures and printed menus into structured prose for /llms-full.txt-style content depth — what AI search engines ground answers on.
- HEIC to JPGiPhone photo onboarding
Convert iPhone-shot dish photos (HEIC) into web-compatible JPG so the rest of the image pipeline can pick them up.
- Image CompressorCore Web Vitals
Compress to AVIF + WebP fallback. Restaurant sites are photo-heavy; getting hero weight down ~70% is the single biggest LCP win you can ship.
- Image ResizeMulti-size from one source
Generate OG images (1200×630), menu cards (800×600), and hero strips (1920×1080) from a single source asset — no manual export passes.
- AI Image GeneratorFestival hero artwork
Brand-palette-matched hero art for festival landings (Diwali, Eid, Christmas, Thanksgiving) without licensing stock photos.
- AI Voice GeneratorVoice search readouts
Reference recordings for Speakable JSON-LD readouts — what Google Assistant, Siri, and Alexa say when a user voice-queries your restaurant.
Case study: Pastry Corner Cafe & Lounge
A Chantilly, Virginia bakery & lounge. 292 menu rows seeded from a single PDF. 53 raw photos compressed to AVIF for ~70% weight reduction. Four festival landings. Voice-search readouts via Speakable JSON-LD. Built entirely with the seven tools above.
Read the case study →FAQ
Do I need to install anything?
No. Everything runs in your browser. The WASM tools (image compression, resize, HEIC conversion, PDF to text) execute locally with no upload. The AI tools dispatch to dedicated GPU workers and notify you when ready.
Is my menu PDF private?
PDF to Text runs client-side — your PDF never leaves your browser. AI Document Summarizer dispatches to a GPU worker; the file auto-deletes within 24 hours and is auditable at /account/tasks.
What does it cost?
Free for the WASM tools. Credit packs cover the AI Studio tools (image generation, document summarizer, voice generator). See pricing — one pack covers a typical restaurant build.