UoR Welcome vs DeCiphr

Your guide to Welcome at the University of Reading. Here you can find the social activities, School timetable, maps, social media, and more.

DeCiphr is an AI-assisted coding tool for linguistic landscape research and other visual data. Photograph signs in the field or import images, apply a coding framework you control, review every result, and export your coded dataset. YOUR FRAMEWORK, APPLIED STRICTLY Define your own coding scheme: codes, definitions, controlled values and a master prompt. Strict coding mode is on by default, so the AI assigns only the values your framework defines and returns "Uncodable" when the image does not support an assignment. Share frameworks as .sigframe files or through the in-app Framework Library. AI CODING AND YOUR INTERPRETATION Split a framework into codes the AI assigns from what is visible and interpretive questions only you answer. The Framework Library includes a two-layer Geosemiotics framework based on Scollon & Wong Scollon (2003). FIELD COLLECTION Take photos in the app, which records the location with your permission, or import photos that carry GPS data. See located items on a map, by project. REVIEW EVERY RESULT Every AI value is editable. Your edits are stored alongside the AI's original values, and both appear side by side in your export. Add analytic memos to each item. EXPORT Export a project, or all projects at once, as an Excel workbook, and export photos as a ZIP numbered to match the rows. PRIVACY Your photos, codings, frameworks and memos are stored on your device. When you run an analysis, that image is sent to Anthropic's Claude API through DeCiphr's relay, which does not keep it. Under Anthropic's API policy, it is not used for training and is deleted within about 30 days, with limited exceptions for legal compliance and abuse monitoring. DeCiphr never stores your photos on its servers. An account is optional, and cloud sync is not yet enabled. Full policy: deciphr.tech/privacy.html FOR RESEARCHERS DeCiphr complements, not replaces, rigorous qualitative research. AI output is a starting point: you review and edit every result and keep full interpretive control. Built by a researcher, for researchers.