Skip to content
Document anonymisation for lawyers

Silüet

Division
03 Products
Status
Beta
Type
Document anonymisation for lawyers
Silüet logo

A desktop app that finds and masks personal data in legal documents (names, ID numbers, addresses, IBANs) entirely on the lawyer’s own computer.

Silüet

Silüet reads petitions, contracts and case files (Word, PDF, UYAP UDF, even scanned pages) and suggests every piece of personal data it finds. The lawyer reviews each suggestion, masks with consistent pseudonyms such as [KİŞİ-1] or [DAVACI], and exports a clean copy. Nothing leaves the computer: no cloud, no upload, no account.

01

The problem

Lawyers share documents all the time: with experts, colleagues, journals and AI tools. Masking names, T.C. Kimlik numbers, addresses and bank details by hand is slow, and one missed line is a KVKK breach. Cloud tools are faster, but they mean uploading the very data you are trying to protect.

02

The product

Silüet runs offline on Mac and Windows. A Turkish-language recognition model and checksum-verified rules detect personal data, while the lawyer stays in control of every decision before anything is exported.

03

Key features

  • 01

    Works offline

    Documents never leave the computer. Case files and originals are stored encrypted on the device.

  • 02

    Built for Turkish legal documents

    DOCX, PDF, UYAP UDF and TXT, plus scanned PDFs and images read with on-device OCR.

  • 03

    Thorough detection

    People, institutions, addresses, T.C. Kimlik and tax numbers (checksum-verified), IBAN, phone, e-mail, plates, dates of birth and case numbers.

  • 04

    Consistent pseudonyms

    The same person gets the same label in every document of a case: numbered [KİŞİ-1] or role-based [DAVACI], [DAVALI].

  • 05

    Review before export

    Every suggestion is accepted or rejected by the lawyer, with the original page shown side by side.

  • 06

    UYAP-ready output

    Masked UDF files open in the UYAP Document Editor, so anonymised filings stay in the format courts use.

04

How it works

  1. 1

    Create a case file

    One file per case or matter keeps pseudonyms consistent.

  2. 2

    Add documents

    Drop in Word, PDF, UDF or scanned files.

  3. 3

    Review suggestions

    Accept, reject or add masks by hand.

  4. 4

    Export

    Save a masked copy, ready to share.

05 — Screenshots

Silüet comparing OCR text with the original page (sample data)
01 — Detections next to the original page
Silüet settings: detection types, confidence threshold and pseudonym format
02 — Choose what to detect and how to label it

Status

Beta

Version 0.1.0 for macOS (Apple Silicon and Intel) and Windows 10/11. The installers are not code-signed yet: on Mac, right-click the app and choose Open; on Windows, choose More info → Run anyway.

Next in Products

Watson

Work with us

Start something with Products.