--- title: Translate Scanned Chinese Docs to English Fast | Smartcat AI description: Translate a scanned Chinese document or photo to English. OCR reads Simplified and Traditional script, AI translates, layout stays. Free 15-day trial. canonical: https://www.smartcat.com/image-translator/translate-scanned-chinese-document/ language: en updated: 2026-08-23 --- # Translate Scanned Chinese Docs to English Fast | Smartcat AI ## Translate Scanned Chinese Documents to English OCR built for Chinese script reads your scan — you choose Simplified or Traditional as the source language, down to the regional variant — and AI translates it into English or any of 280+ languages, with the layout intact. One limit worth knowing before you upload: this handles horizontally set Chinese. Vertical, right-to-left columns, common in Taiwanese and older mainland documents, are not supported. - [Book a demo](https://www.smartcat.com/book-a-demo/?n-upl) - [Sign up free](https://smartcat.com/sign-up?n-upl) ## How does Chinese scanned-document translation work? ### Upload the scan A scanner file or a straight-on phone photo of the page. Scan at 300 DPI or better, keep the page straight, and light it evenly with good ink-to-paper contrast. Chinese characters are less forgiving of low resolution than Latin text, because the difference between two characters can be a single short stroke. ![Upload the scan](https://gqucouwbfxremfdhdosu.supabase.co/storage/v1/object/public/sc-store/1733694209742Translate_documents_in_five_simple_steps01.webp) ### Set the source language yourself Smartcat can detect the source language, but on Chinese it is worth **setting it by hand**. Chinese (simplified) and Chinese (traditional) are distinct selectable languages: - **Simplified** — mainland China, Singapore and Malaysia. - **Traditional** — Taiwan, Hong Kong and Macau. Each has regional variants, plus **Cantonese and Pinyin**. The variant you choose is what recognition and translation actually run against. Then set English, or any of **280+ languages**, as the target. ![Set the source language yourself](https://gqucouwbfxremfdhdosu.supabase.co/storage/v1/object/public/sc-store/1733694222040Translate_documents_in_five_simple_steps02.webp) ### Recognition, segmentation, then translation Optical character recognition trained on Chinese script converts the scanned image into machine-readable text. Because written Chinese has no spaces between words, that text is then segmented into words before translation — wrong segmentation produces wrong meaning even when every character was read correctly. Only then is the segmented text translated. ![Recognition, segmentation, then translation](https://gqucouwbfxremfdhdosu.supabase.co/storage/v1/object/public/sc-store/1733694236073Translate_documents_in_five_simple_steps03.webp) ### Review what the OCR actually read The extracted Chinese stays visible and editable alongside the translation, so a bilingual colleague or a [Marketplace](https://www.smartcat.com/marketplace/) reviewer can check the source rather than the output. That is the step that catches visually similar characters — 未 and 末, 己 and 已 — misread on a low-quality scan, which otherwise produces fluent, confident, wrong English no monolingual reader could spot. ![Review what the OCR actually read](https://gqucouwbfxremfdhdosu.supabase.co/storage/v1/object/public/sc-store/1733694258982Translate_documents_in_five_simple_steps04.webp) ### Download the English document The source layout is preserved, which is what makes side-by-side comparison possible: clause 4.2 in English sits where 第4.2条 sits in the source. English translations of Chinese typically run longer than the source, so translated text boxes stay editable for fitting. ![Download the English document](https://gqucouwbfxremfdhdosu.supabase.co/storage/v1/object/public/sc-store/1733442385341Image_01.webp) Ready to translate a scanned Chinese document? - [Start My Free Trial](https://smartcat.com/sign-up?main-12-translate-document) ## One workflow for Chinese scans — and every other file you send This page is for image scans: JPG, PNG and phone photos. A Chinese PDF, including a scanned one, belongs with the [Chinese-English PDF translator](https://www.smartcat.com/chinese-english-pdf-translator/) — choose the *pdfimageonly* parsing method at upload and each page runs through this same recognition, multi-page files included. For scans in other languages, start at [translate scanned documents](https://www.smartcat.com/image-translator/translate-scanned-document/). Every other format has its own workflow too. ### PDF translator Scanned PDFs use the *pdfimageonly* parsing method, which runs this same OCR page by page and keeps the original layout. ### Google Docs translator Ideal for quick edits to Google Docs files. ### Google Sheets translator Handle data-heavy files without losing structure. ### Google Slides translator Keep presentations consistent across languages. ### PowerPoint translator Maintain visuals and text alignment across slides. ### Excel translator Translate your spreadsheets efficiently. ### XLIFF translator Translate XLIFF files for professional localization workflows. ### TXT file translator Translate your TXT files with Smartcat easily. ### HTML translator Translate HTML content for your website. ### XML translator A simple XML translator for structured data. ### Image translator Translate the text inside images while keeping the original layout. ### JPG translator Translate JPG files with Smartcat's image translation tool. ### PNG translator Translate PNG files with Smartcat's image translation tool. ### SRT translator Translate SRT files and make subtitles for global audiences. ### VTT translator Upload your VTT files and translate them now. ### MP3 translator Instantly translate your audio into multiple languages. ### MP4 translator Translate your video files for global audiences. ### WAV file translator Record and translate WAV files into your preferred languages. ### MOV file translator Translate MOV files with Smartcat's MOV file translator. ### GIF translator Translate the text in a GIF — for an animated GIF, the first frame is translated. ## Rated by the people who use it ### 9.6/10 for ease of setup ### 9.3/10 for ease of use ### 1,000+ enterprise brands ### 4.6/5 average rating on G2 See How Your Team Can Benefit A 1:1 walkthrough with our team, no commitment — bring the Chinese documents you actually need read. - [Book a Personalized Demo](https://www.smartcat.com/book-a-demo/?main-12-translate-document) ## What parts of a Chinese document can OCR actually read? ### 300 DPI ### the input bar for hanzi Scan at 300 DPI or better, page straight, evenly lit, high in contrast. Printed Simplified and printed Traditional text both translate cleanly on a decent scan; below the bar, character edges blur and recognition starts guessing. ### 2 scripts ### Simplified and Traditional, each selectable With regional variants for the PRC, Singapore, Malaysia, Taiwan, Hong Kong and Macau, plus Cantonese and Pinyin. Mixed Chinese-English pages are recognized in one pass, and dates written 年 / 月 / 日 are converted to the target-language convention. ### Vertical ### right-to-left columns: not supported **Vertical, right-to-left columns** — common in Taiwanese and older documents — are not supported. Rotate to a horizontal source where you have one, or hire a Chinese-English translator from the [Marketplace](https://www.smartcat.com/marketplace/). Two more things the machine cannot read: - **Red seals and chops** — usually carved in stylized seal script, so they stay in the layout, but a human has to read them. - **Handwritten characters** — unreliable, and cursive hands need a human translator. ## What teams get back at platform level ### $1.20 ### Per 1,000 words, down from $200–$300 Stanley Black & Decker, across 40+ countries in 6 languages, moved L&D translation to $1.20 per 1,000 words and cut average turnaround by two weeks. - [Read case study](https://www.smartcat.com/cases/stanley-black-and-decker/) ### 31 hours ### Of work time saved monthly for Babbel’s marketing and L&D departments - [Read case study](https://www.smartcat.com/cases/babbel/) ### 3 days ### to deliver what took 10 Smith+Nephew cut translation turnaround from an average of 10 days with previous providers to two to three days for the same content. - [Read case study](https://www.smartcat.com/cases/smith-nephew/) From a Chinese scan to English you can act on Upload the page, get an English document that mirrors it, and check the extracted Chinese with anyone who reads it. Free for 15 days with 15,000 Smartwords, no credit card, with full access to translation capabilities. - [AI-Translate a File Now](https://smartcat.com/sign-up?main-f-translate-document) ## Frequently Asked Questions ### How do I translate a scanned Chinese document to English? Upload the scan or a phone photo of the page and set the source language — Chinese (simplified) or Chinese (traditional), including regional variants such as Taiwan, Hong Kong, Macau, Singapore and Malaysia. Character recognition extracts the text, it is segmented into words, and AI translates it into English while keeping the document’s layout. Review the result and download an English document that mirrors the original. ### Do I need to know whether my document is Simplified or Traditional? You have both options — Smartcat can detect the source language, or you can set it yourself per file — but on Chinese, **set it yourself**. Simplified and Traditional are **separate source languages**, each with regional variants, so you choose when you set up the job. It is usually easy to tell by eye; if you are unsure, ask anyone who reads Chinese before you upload. The variant you select is what recognition and translation run against. ### Does it handle vertical Chinese text? No. Vertical, right-to-left columns are not supported today. Rotate to a horizontal source where you can, or route vertically set documents to a human translator — you can hire a Chinese-English one from the [Marketplace](https://www.smartcat.com/marketplace/) inside the same project. ### Can I translate a photo of a Chinese document taken with my phone? Yes — shoot it straight-on in even light and it behaves like a scan. Skewed or shadowed photos multiply recognition errors in hanzi faster than in Latin text, so retaking the photo flat is always worth it. ### How do I know the translation is right if I can’t read Chinese? You cannot verify it by rereading the English, so the workflow gives you **two checks that do not depend on you**. - The **extracted Chinese text is visible and editable**, so any bilingual colleague can confirm what the OCR read in seconds. - For documents with stakes, add a professional Chinese-English reviewer from the [Marketplace](https://www.smartcat.com/marketplace/) inside the same project. AI output on a clean scan is a strong draft; it is not self-verifying. ### Can it read the red seals and stamps? Mostly no — official chops are typically stylized seal script, which is a different writing system from printed text. The seal stays in place in your translated document, but have a human read what it says. ### What about handwritten Chinese? Printed text translates reliably; handwriting does not. Neat block characters sometimes extract, cursive and grass-script hands essentially never. For a handwritten document, hire a human translator — the [Marketplace](https://www.smartcat.com/marketplace/) is the shortest path to one. ### Will the translated document keep the original formatting? Yes — layout preservation is what makes side-by-side comparison possible, so clause 4.2 in English sits where 第4.2条 sits in the source. English translations of Chinese typically run longer than the source text, so translated text boxes stay editable for fitting. ### What file types can I upload? JPG/JPEG/JFIF, PNG, TIF/TIFF, BMP, GIF, DCX/PCX, JP2/JPC, DJVU/DJV, JB2, image-only PDF and AI files are processed via OCR, up to 30 MB per image. Single-page TIFF is explicitly supported. HEIC and WebP are not on the supported list, so convert those first. ### Should I use this page or the Chinese PDF translator? Image files and photos — this page. PDF files, including scanned ones — the [Chinese-English PDF translator](https://www.smartcat.com/chinese-english-pdf-translator/). Same recognition, different container. For scans in languages other than Chinese, start at [translate scanned documents](https://www.smartcat.com/image-translator/translate-scanned-document/). ### Is my document confidential? The documents people upload here — contracts, licenses, records — are usually the sensitive kind. Smartcat is SOC 2 Type II compliant; files are encrypted in transit and at rest, workspaces are isolated, and access is role-based. Details on the [security page](https://www.smartcat.com/security/). ### Is there a free way to translate a scanned Chinese document? Free for 15 days with 15,000 Smartwords and no credit card, with full access to translation capabilities. Upload a scan, set the Chinese variant, translate it into English and download the result to judge the quality yourself. After the trial it is a paid plan. ## Sources - Statista Market Insights. "[Language translation NLP – Worldwide.](https://www.statista.com/outlook/tmo/artificial-intelligence/natural-language-processing/language-translation-nlp/worldwide)" Statista, March 2025. - Statista Research Department. "[Number of employees in interpreting and translating services in the United States from 2011 to 2022.](https://www.statista.com/statistics/320340/number-of-employees-in-interpreting-and-translating-services-us/)" Statista, 2023. - Statista Research Department. "[Market size of translation services in the United States from 2015 to 2022.](https://www.statista.com/statistics/1280441/translation-services-market-size-usa/)" Statista, 2023. - Liu, Z., Wu, C., & Xu, X. (2025). The Role of Artificial Intelligence in Sustainable Development and Industrial Transformation. [*Asia Pacific Economic and Management Review*](https://doi.org/10.62177/apemr.v2i2.185), *2*(2).