Tamil to English Document Translation API
Yes. The LandLens translation API translates Tamil Nadu property documents between Tamil and English: sale deeds, pattas, chitta and TSLR extracts, encumbrance certificates and notices. Send a public link to the PDF or scan; it is read (scans included) and translated, and you get the translation as text and as structure: the issuing office, the title, every labelled field, every table with its columns, and the running text. Survey, patta and document numbers, dates and amounts are kept exactly as printed, and names keep their Tamil form in brackets. It costs 3 credits per document, and nothing when the document cannot be read.
https://verified.realestate/api/v1/tools/translate_document- Price
- 3 credits per answer
- Source
- Your document, read and translated
- Coverage
- Tamil Nadu property documents, Tamil and English
- Answers in
- Usually 15 to 20 seconds
What it returns
One call, one answer. found is false (status not_found) when the document could not be fetched or read, and then nothing is charged.
| Field | What it holds |
|---|---|
translation | The whole translation as plain text. |
structured.from, to | The languages translated from and into. |
structured.issuer, title | Who issued the document, line by line, and its own title. |
structured.fields[] | Every labelled value the document prints (district, village, patta number, survey number, owners, document number, dates, office, extent), label and value, in order. |
structured.tables[] | Every table with its title, translated columns and rows. |
structured.sections[] | The running text, by heading and paragraph. |
structured.unreadable, note | Fragments that could not be read or placed, and a line when something affected the translation (a cut page, a stamp over the text). |
direction, source_file, characters | The direction used, the file name read, and the length of the translation. |
Inputs
The JSON body of the POST is the arguments. Nothing is required by the schema, but over the API file_url is: without it the call answers not_found and charges nothing.
| Argument | Type | What to send |
|---|---|---|
file_url |
string | A public https link to the PDF or scan, up to 25 MB. Links into a private network are refused. |
direction |
string: tamil-to-english, english-to-tamil, auto-detect | "tamil-to-english", "english-to-tamil" or "auto-detect" (the default). |
Code
Set LANDLENS_KEY to your key. In an AI app, connect https://verified.realestate/mcp once and ask in plain words instead. In Postman, download this API's collection, import it (File, Import) and set apiKey.
curl -X POST https://verified.realestate/api/v1/tools/translate_document \
-H "Authorization: Bearer $LANDLENS_KEY" \
-H "Content-Type: application/json" \
-d '{}'
import os, requests
r = requests.post(
"https://verified.realestate/api/v1/tools/translate_document",
headers={"Authorization": f"Bearer {os.environ['LANDLENS_KEY']}"},
json={},
timeout=120,
)
body = r.json()
print(body["status"], body["credits_charged"], body["data"])
const res = await fetch("https://verified.realestate/api/v1/tools/translate_document", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.LANDLENS_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({}),
});
const body = await res.json();
console.log(body.status, body.credits_charged, body.data);
Translate this patta into English and tell me who the owners are and what the land classification is: https://example.com/patta.pdf
Your AI fetches the document, translates it, and reads the owners and classification from the translated fields.
Responses
Every answer has the same envelope. What status can say for this call:
| status | What it means |
|---|---|
ok | The document was read and translated. Charged. |
not_found | No file_url, a link that could not be fetched or is not a document, text too short to translate, or the translation service did not answer. error says which. Not charged. |
insufficient_credits | HTTP 402, with the credits needed, your balance and where to top up. |
error | The call failed on our side. Nothing is charged. |
Where the data comes from
Not a register: the document you send, read page by page (including scanned images) and translated by an AI model instructed for Tamil Nadu legal, revenue and property documents.
PDF and scan images (JPEG, PNG, TIFF, BMP, WebP, HEIC) up to 25 MB, and up to 50,000 characters of text per call. Tamil to English, English to Tamil, or detected from the document.
Not applicable. Each call translates the document you send at that moment.
The document is fetched from your link and its text read, then translated into a fixed layout (issuer, title, fields, tables, sections). Legal and revenue terms are translated with their meaning in brackets the first time (Punjai (dry land)), and a word that cannot be read with confidence is written [unclear], never guessed.
Who uses it
Banks and NBFCs
Read Tamil title documents at login without waiting for a translator, with every number kept as printed.
NRI buyers and their advisers
Understand a Tamil sale deed or patta before signing, laid out as fields and tables.
Document platforms
Turn uploaded Tamil records into structured English fields your own system can read.
Limits and good practice
- This is an AI translation. Court and registration use still needs a certified translation.
- Up to 50,000 characters of text are translated in one call; a longer document is cut there. Send long deeds in parts.
- The link must be publicly reachable and point at a PDF or an image. A web page, a private link or a file over 25 MB is refused.
- A word the reader cannot make out is marked [unclear]. A poor scan gives a poorer translation; send the clearest copy you have.
Pricing
A translation costs 3 credits per document, which covers fetching and reading it. A document that cannot be fetched or read, or a translation that does not complete, is not charged. Credits come from the same LandLens wallet as the chat and the minis. How billing works.
For thousands of calls a month, a portfolio sweep or a platform built on these registers, we quote a customised bulk price, with invoicing to match.
Questions
Is there an API to translate Tamil land documents into English?
Yes. LandLens translates Tamil Nadu property documents (deeds, pattas, ECs, notices) between Tamil and English from a link to the PDF or scan, and returns the translation as text and as structured fields and tables.
Does it work on scanned documents?
Yes. Scans and photos (JPEG, PNG, TIFF, BMP, WebP, HEIC) are read before they are translated.
How much does it cost?
It costs 3 credits per document over the API, which includes reading it. Nothing is charged when the document cannot be read.
Can I use the translation in court or at the registrar?
No. It is an AI translation for understanding the document. Official use needs a certified translation.
Are names and numbers changed?
Survey, patta and document numbers, dates and amounts are kept exactly as printed. Names are written in English letters with the Tamil form in brackets.
Updated 22 Oct 2026 · OpenAPI description