Doc Translator

What is Doc Translator?

Doc Translator is a web-based AI translation service designed around files rather than pasted text. Instead of returning only a block of translated copy, it aims to give back a finished document that looks like the original: columns, tables, fonts, images and charts are intended to stay where they were, so the translated file can be shared without reformatting.

The product page positions it as "an AI document translator built for real work - fast, accurate, and faithful to the file you started with." It states support for 150+ languages and more than 50 file types, and claims a user base of 2M+. A free entry point exists ("Free to start" and "Sign in and translate for free" appear on the sign-in flow), with paid tiers for higher volume.

How Doc Translator works

A user drops a file into the upload area on the homepage or selects it manually, then sets a source language (Auto Detect is available) and a target language before starting the job. The same page also offers dedicated entry points for PDF, EPUB and image translation.

Billing is credit-based. After a file is uploaded, the service analyzes it and shows an estimated credit cost before the user confirms; the exact amount is recalculated and reserved when the translation task is created. Successful jobs consume the reserved credits, and failed jobs are refunded automatically.

Text extraction differs by file type. Word, PowerPoint, Excel, EPUB, subtitle and structured files use extracted source text, as do PDFs with selectable text. Scanned, image-only, mixed or unclassified PDFs are routed through DeepL OCR, which is priced at a much higher rate.

Main features

• Formatting preservation. The stated goal is pixel-perfect output, with columns, tables, fonts, images and charts unchanged.
• Broad format support. PDF, PPTX, DOC, DOCX, XLS, XLSX, CSV, JSON, XML, YAML, EPUB, VTT, SRT, JPG, PNG, WEBP, MD and TXT.
• Language coverage. 150+ languages with an auto-detect source option.
• OCR for scanned documents. Image-only and mixed PDFs are handled through OCR rather than rejected.
• Subtitle and image translation. VTT and SRT files, plus JPG, PNG and WEBP images, are supported natively.
• Credit transparency. An estimate is shown before each job, and failed jobs are refunded.
• Tiered throughput. Higher plans raise upload size, concurrent job limits and processing priority.

Pricing

Three plans are listed, all shown with annual billing figures and a "Monthly / Yearly" toggle advertising 50% off on yearly billing. The monthly-equivalent rates themselves are not stated on the pages reviewed.

• Basic - $4.90/month (billed annually, stated as saving about $60 vs. monthly): 3,000 credits per month, uploads up to 10 MB, 2 concurrent translation jobs, standard processing and support.
• Pro - $9.90/month (billed annually, stated as saving about $120 vs. monthly), labelled most popular: 10,000 credits per month, uploads up to 25 MB, 5 concurrent jobs, priority processing and support.
• Max - $19.90/month (billed annually, stated as saving about $240 vs. monthly): 25,000 credits per month, uploads up to 50 MB, 10 concurrent jobs, highest processing priority and support.

Every paid tier includes the same core translation tools, with limits scaling by plan. Credit rates are: text translation at 1 credit per 2,000 characters (rounded up, minimum 1 credit); document translation at 1 credit per page; text-based PDF translation at 1 credit per page; scanned PDF translation at 100 credits per analyzed page with a minimum of 1,500 credits per PDF; and image translation at 60 credits per image for the current medium-quality GPT Image edit. A standard page is estimated at roughly 2,000 source characters, so 1 credit is roughly 1 page.

Common use cases

The credit model and format coverage point to several recurring scenarios: legal and contractual documents that must retain clause structure; localization and marketing teams translating decks, one-pagers and product documentation; students and researchers working with academic PDFs and EPUBs; subtitle translation for video content using SRT and VTT; structured-data files such as JSON, XML, YAML and CSV that need their keys and hierarchy intact; and image or scanned paperwork that requires OCR before translation.

Limitations

• Credits, not unlimited usage. Volume is metered, and heavy users can exhaust a monthly allowance before the cycle ends.
• Scanned PDFs are expensive. At 100 credits per analyzed page and a 1,500-credit minimum per PDF, a single scanned file can consume half of the Basic plan's monthly 3,000 credits.
• Upload ceilings. Files are capped at 10 MB (Basic), 25 MB (Pro) and 50 MB (Max).
• Concurrency limits. Two, five or ten simultaneous jobs depending on tier.
• Thin free-tier detail. "Free to start" is advertised, but the supplied pages do not document what the free allowance includes.
• No API or enterprise plan is described on the pages reviewed, and no desktop or offline option is mentioned.
• Estimates vary. Costs are derived from source-character counts, so the final charge depends on the actual text extracted.
• Data handling specifics are not covered beyond links to the Privacy Policy and Terms of Use.

Summary

Doc Translator targets the specific pain point of translating documents without destroying their formatting. Its differentiators are breadth of accepted file types, 150+ language coverage and layout fidelity, while its main trade-off is a credit-based cost model where oversized or scanned inputs can get expensive quickly. For occasional PDFs, Office files, EPUBs, subtitles or images, the entry tier is inexpensive; for scanned archives or high-volume workflows, the higher plans or careful sizing of source files matter more.