Syaahi
Offline-First Document Studio for Android

Your scans, Hindi & English OCR, and PDFs—strictly on your device.

Syaahi combines a CameraX perspective scanner, bilingual Devanagari & Latin text recognition, Unicode searchable PDF compilation, batch image studio, and QR/ZIP utilities without forcing accounts or uploading your documents.

Download for AndroidExplore CapabilitiesGoogle Play Listing

dev.syaahi.app • Android 8.0+ (API 26–35) • Zero INTERNET permission in Android v1.0

Nibi, Syaahi's original Ink and Paper mascot holding an obsidian ink-nib crest
Meet NibiStudio Companion
Syaahi's calm companion guiding on-device scanning, Hindi & English OCR, and clean document workflows.
Zero INTERNET Permission• Enforced via tools:node="remove" in Android manifest
Offline Hindi & English OCR• Bundled Google ML Kit Latin + Devanagari models
CameraX 4-Corner Warp• Precision magnifier loupe & homography correction
Unicode Searchable PDFs• Selectable Noto Sans Devanagari text layer
On-Device Room Workflows• 100% local database with zero forced accounts
Magic Boost & B&W Filters• Real-time paper scan enhancement & shadow cleanup
Biometric App Lock & Vault• Native Android BiometricPrompt security
PDF Merge, Split & Compress• Zero server uploads—processed in local device RAM
ZXing Offline QR Studio• Private scannable vector generation & scanner
Four Document Engines

Built for real paperwork—not bloated subscriptions

Every core transformation executes cleanly in on-device RAM and local Android storage.

Perspective Scanner

Full-resolution CameraX capture, automatic EXIF orientation handling, document edge detection, four-corner loupe alignment, and homography warp.

Hindi & English OCR

Runs bundled Google ML Kit Latin and Devanagari recognizers offline, embedding selectable Noto Sans Devanagari Unicode text layers inside PDFs.

PDF Toolkit

Merge multi-document packs, split page ranges, compress oversized files, convert images to PDF, render pages back to images, and stamp clean watermarks.

Image & Batch Studio

Crop, rotate, flip, sharpen scans with Magic Boost or High-Contrast B&W, apply smart resolution/quality presets, and export ZIP archives or QR codes.

Interactive Bilingual OCR Simulator

Try Google ML Kit Devanagari & Latin Extraction

Hover over any text block to inspect on-device ML Kit recognition bounds and confidence. Click any block to copy its clean Unicode text directly to your clipboard.

Sample:
SYAAHI ON-DEVICE OCR • BILINGUAL TEXT LAYER
Google ML Kit v16.0 • Local RAM
कार्यालय आदेश / Office MemorandumBilingual ML Kit • 99.9%
दस्तावेज़ क्रमांक: SY-2026-8819 • दिनांक: 15 अगस्त 2026Devanagari ML Kit • 99.6%
प्रमाणित किया जाता है कि सभी दस्तावेज़ और पाठ्य निष्कर्षण केवल उपकरण पर ही संसाधित होते हैं।Devanagari ML Kit • 99.8%
CameraX 4-corner perspective warp applied with 100% offline ML Kit Latin recognizer.Latin ML Kit • 99.9%
देवनागरी एवं लैटिन अक्षरों का सटीक एवं त्वरित निष्कर्षण Noto Sans Devanagari यूनिकोड परत के साथ।Devanagari ML Kit • 99.7%
Zero INTERNET Permission Enforced • tools:node="remove" • Local RAM Processing OnlyLatin ML Kit • 100%
सत्यापित डिजिटल मुहर: SYAAHI OFFLINE VERIFIED • NIBI SEAL OF TRUSTBilingual ML Kit • 99.5%
🎯 Embedded UTF-8 Noto Sans Devanagari text layer ready for PDF compilation.Tap text block to copy
Verifiable Offline Guarantees

Offline-first isn't a slogan—it's enforced in the manifest.

Syaahi strips android.permission.INTERNET during Android manifest merging (tools:node="remove") so even transitive SDK telemetry transports remain disabled at the OS layer in v1.0.

Whenever optional cloud accounts, remote announcements, or this website's support form are used, we document every collected field, encryption standard, and deletion procedure openly—never making misleading blanket claims.

Platform Data Boundary

  • Scanned pages & OCR text: Processed locally on-device in RAM and exported to Downloads/Syaahi. Never uploaded.
  • Workspace defaults: Anonymous local mode active immediately on launch; zero mandatory sign-up or email walls.
  • Optional security lock: System BiometricPrompt gating configurable in Settings.
  • Optional cloud/support interaction: Plainly disclosed in Privacy & Account Deletion documentation.
Authentic Product Interface

Warm Ivory by day, Deep Ink by night

Designed around the Ink × Paper Material 3 system paired with genuine device captures from Syaahi for Android.

Split: 50%
Syaahi Warm Ivory light theme showcase: Home Dashboard and Hindi/English Searchable PDF OCR Workspace
Syaahi Deep Ink dark theme showcase: Edge Detection Scanner with Loupe Magnifier and Nibi Empty State
Warm Ivory (Light Mode)
Deep Ink (Obsidian Dark)
Drag the handle or use arrow keys to inspect day vs. night theme styling.Click or drag anywhere • Keyboard Arrow keys enabled
Actual Syaahi Android Home screen capture showing Scan Document, Image to PDF, Compress PDF, Extract Text, Merge PDF, and Resize Image actions
Device capture — Home quick-action grid on Android 15 (Galaxy A14 5G).
Actual Syaahi Android Image Editor showing Crop, Resize, Compress, and Enhancement Filters on scanned Devanagari print
Device capture — Image Editor enhancement filters & crop studio.
Actual Syaahi Android Interactive Scannable QR Studio generating offline QR code bitmap
Device capture — ZXing Interactive Offline QR Studio.
Bundled Ecosystem

Everything included—Explore More directory

Zero third-party ad networks. Launch integrated tools directly inside Syaahi on Android.

View All Workflows & Deep Links
On-Device AIOffline Ready

Hindi & English OCR Engine

On-device Devanagari & Latin recognition

Extract editable, copyable text from printed bills, receipts, notices, and forms using bundled ML Kit recognizers.

syaahi://ocr
Document ToolsZero Cloud

Searchable PDF Compiler

Unicode selectable text layer inside scanned PDFs

Embed Hindi Devanagari and English OCR text into PDFs with bundled Noto Sans Devanagari fonts via PDFBox-Android.

syaahi://documents
Built-in WorkflowIncluded

Batch Image Studio

Multi-image resize, compress & format export

Crop, rotate, flip, enhance contrast, apply custom width caps, and export batches to JPEG or PNG offline.

syaahi://batch-studio
UtilitiesIncluded

QR, ZIP & SHA-256 Toolkit

Offline archive packing & integrity verification

Verify downloaded assets with SHA-256 digests, package multi-file ZIP archives with SAF, and generate crisp ZXing QR codes.

syaahi://utilities
AutomationLocal First

Saved Toolchains

Repeatable local document pipelines

Keep saved presets for scan-to-PDF and image optimization cleanly inside your on-device Room database.

syaahi://workflow
Frequently Asked Questions

Honest answers, clear technical facts

Yes. The current Android build (dev.syaahi.app v1.0) strips android.permission.INTERNET at build time using tools:node="remove". Document scanning, Hindi/English OCR, searchable PDFs, ZIP creation, QR codes, and image filters execute purely on your device.

Transparency is critical. While the current v1.0 APK runs offline with zero network calls, the Syaahi ecosystem also includes this website (with a /contact feedback form) and optional backend infrastructure built for future opt-in releases (optional accounts, remote config, and Firebase Cloud Messaging announcements). Our Privacy Policy discloses every data path honestly so we never make misleading "zero data everywhere" claims.

Yes. Syaahi bundles Google ML Kit Devanagari and Latin recognizers alongside Noto Sans Devanagari TrueType font embedding inside PDFBox-Android so Hindi matras, conjuncts, and English words remain selectable and searchable in standard PDF readers.

No. Anonymous local workspace mode is permanent and default. Even on future builds with optional account support, scanning, OCR, PDF assembly, ZIP archiving, and image editing are never gated behind login.

Exported PDFs, images, ZIPs, and QR graphics are saved inside Downloads/Syaahi (or app-private storage fallback if Downloads writes conflict). You can reset local Room history and preferences inside Account Settings -> Cancel Local Profile & Clear History, or request deletion of optional cloud/feedback records via /account-deletion.