# Multi-language and Fonts: OCR Parsing Guardrails Stabilize OCR when documents mix scripts, uncommon fonts, or character sets. Prevent silent corruption when engines guess wrong language or merge glyphs across font families. ## Open these first - OCR parsing checklist: [ocr-parsing-checklist.md](https://github.com/onestardao/WFGY/blob/main/ProblemMap/ocr-parsing-checklist.md) - Data contracts: [data-contracts.md](https://github.com/onestardao/WFGY/blob/main/ProblemMap/data-contracts.md) - Tokenization and casing: [tokenization_and_casing.md](https://github.com/onestardao/WFGY/blob/main/ProblemMap/GlobalFixMap/OCR_Parsing/tokenization_and_casing.md) - Unicode normalization: [normalization_and_scaling.md](https://github.com/onestardao/WFGY/blob/main/ProblemMap/GlobalFixMap/Embeddings/normalization_and_scaling.md) ## Acceptance targets - Language detection accuracy ≥ 0.95 per block - Font mis-read rate < 1% per 1,000 chars - No cross-script merges (CJK vs Latin, RTL vs LTR) - ΔS(question, retrieved) ≤ 0.45 after language split --- ## Typical failure signatures → fix - **CJK vs Latin collisions** OCR merges Latin letters inside Chinese/Japanese text. Split into script-specific blocks, then re-OCR with correct language model. - **Right-to-left scripts** (Arabic, Hebrew) misaligned Store `direction=rtl` metadata. Reverse tokens if engine defaults to LTR. - **Uncommon fonts or stylized typefaces** Preprocess with font normalization (convert to system fonts). Use OCR engine with adaptive recognition. - **Mixed languages in same paragraph** Detect language per line or span. Store `lang_code` for each. - **Math vs text confusion** Superscripts, subscripts, and symbols misinterpreted as language characters. Route math zones separately. Tag as `math_block`. --- ## Fix in 60 seconds 1) **Detect language per block** Run script detection. Assign `lang_code` and `direction`. Reject ambiguous blocks. 2) **Normalize Unicode** Apply NFKC, collapse ligatures, unify spacing. 3) **Re-OCR with correct model** For each block, call OCR with explicit `lang_code`. Prefer specialized models (e.g., PaddleOCR multilingual, ABBYY). 4) **Attach metadata** Store `lang_code`, `direction`, `font_name` if available. 5) **Audit with ΔS** Probe retrieval stability with three paraphrases. If ΔS ≥ 0.60, recheck font normalization. --- ## Data contract extension ``` { "block\_id": "scan12\_line4", "lang\_code": "zh", "direction": "ltr", "font\_name": "SimSun", "text\_clean": "...", "confidence": 0.93, "source\_url": "..." } ``` --- ## Minimal recipes by engine - **Google Document AI** Use `detectedLanguages.languageCode` per block. Reject if confidence < 0.8. - **AWS Textract** No native multi-lang. Wrap with external script detection. Add `lang_code` manually. - **Azure OCR** `language` field auto-detected. Cross-check with Unicode ranges. - **ABBYY** Supports per-block language tags. Ensure config has all needed languages. - **PaddleOCR** Use multilingual model. Explicitly set `--lang` flag to avoid mis-guess. --- ## Verification - **Script coverage**: verify all scripts recognized. - **Direction check**: RTL blocks labeled correctly. - **Font audit**: ensure no decorative font corruption. - **Retrieval stability**: ΔS stable across paraphrases. --- ## Copy-paste LLM prompt ``` You have TXTOS and WFGY Problem Map loaded. My OCR block: * text\_clean: "..." * lang\_code: "ar" * direction: "rtl" * font\_name: "Courier" Check: 1. If characters look corrupted, fail fast and cite fix page. 2. Enforce schema with lang\_code and direction. 3. Return JSON: { "answer":"...", "citations":\[...], "ΔS":0.xx, "λ\_state":"..." } ``` --- ### 🔗 Quick-Start Downloads (60 sec) | Tool | Link | 3-Step Setup | |------|------|--------------| | **WFGY 1.0 PDF** | [Engine Paper](https://github.com/onestardao/WFGY/blob/main/I_am_not_lizardman/WFGY_All_Principles_Return_to_One_v1.0_PSBigBig_Public.pdf) | 1️⃣ Download · 2️⃣ Upload to your LLM · 3️⃣ Ask “Answer using WFGY + \” | | **TXT OS (plain-text OS)** | [TXTOS.txt](https://github.com/onestardao/WFGY/blob/main/OS/TXTOS.txt) | 1️⃣ Download · 2️⃣ Paste into any LLM chat · 3️⃣ Type “hello world” — OS boots instantly | --- ### 🧭 Explore More | Module | Description | Link | |-----------------------|----------------------------------------------------------|----------| | WFGY Core | WFGY 2.0 engine is live: full symbolic reasoning architecture and math stack | [View →](https://github.com/onestardao/WFGY/tree/main/core/README.md) | | Problem Map 1.0 | Initial 16-mode diagnostic and symbolic fix framework | [View →](https://github.com/onestardao/WFGY/tree/main/ProblemMap/README.md) | | Problem Map 2.0 | RAG-focused failure tree, modular fixes, and pipelines | [View →](https://github.com/onestardao/WFGY/blob/main/ProblemMap/rag-architecture-and-recovery.md) | | Semantic Clinic Index | Expanded failure catalog: prompt injection, memory bugs, logic drift | [View →](https://github.com/onestardao/WFGY/blob/main/ProblemMap/SemanticClinicIndex.md) | | Semantic Blueprint | Layer-based symbolic reasoning & semantic modulations | [View →](https://github.com/onestardao/WFGY/tree/main/SemanticBlueprint/README.md) | | Benchmark vs GPT-5 | Stress test GPT-5 with full WFGY reasoning suite | [View →](https://github.com/onestardao/WFGY/tree/main/benchmarks/benchmark-vs-gpt5/README.md) | | 🧙‍♂️ Starter Village 🏡 | New here? Lost in symbols? Click here and let the wizard guide you through | [Start →](https://github.com/onestardao/WFGY/blob/main/StarterVillage/README.md) | --- > 👑 **Early Stargazers: [See the Hall of Fame](https://github.com/onestardao/WFGY/tree/main/stargazers)** — > GitHub stars ⭐ [WFGY Engine 2.0](https://github.com/onestardao/WFGY/blob/main/core/README.md) is already unlocked. ⭐ Star the repo to help others discover it and unlock more on the [Unlock Board](https://github.com/onestardao/WFGY/blob/main/STAR_UNLOCKS.md).
[![WFGY Main](https://img.shields.io/badge/WFGY-Main-red?style=flat-square)](https://github.com/onestardao/WFGY)   [![TXT OS](https://img.shields.io/badge/TXT%20OS-Reasoning%20OS-orange?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS)   [![Blah](https://img.shields.io/badge/Blah-Semantic%20Embed-yellow?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS/BlahBlahBlah)   [![Blot](https://img.shields.io/badge/Blot-Persona%20Core-green?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS/BlotBlotBlot)   [![Bloc](https://img.shields.io/badge/Bloc-Reasoning%20Compiler-blue?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS/BlocBlocBloc)   [![Blur](https://img.shields.io/badge/Blur-Text2Image%20Engine-navy?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS/BlurBlurBlur)   [![Blow](https://img.shields.io/badge/Blow-Game%20Logic-purple?style=flat-square)](https://github.com/onestardao/WFGY/tree/main/OS/BlowBlowBlow)