Your screen changes.
You get notified.

OKYA Watch is a browser-based screen recognition and automation product by OKYA. After you share a screen, it locally recognizes text in your selected language, numbers, colors and barcodes, then runs your configured rules for alerts, extraction, recording or translation.

No installation is needed. Sign in to your OKYA account to use the app. All features are currently open to members. You must authorize sharing a window, tab or screen before recognition starts.

Add a source in OKYA Watch and authorize the window or tab you need. Create a text recognition task, select the text region and OCR language, then save and start recognition. Copy the result or export recognition records as CSV. It reads visible text in the shared image, not webpage source code or document files.

Create a text task, select a region and choose either text changed or contains keyword. Text changed compares successive recognition results; a keyword can be Error, Done or any text you want to watch. A match can trigger an in-page or system browser notification. System notifications require permission.

Yes. Select a region containing one number, then choose number changed, greater than or less than. For example, get an alert when stock falls below 10 or a price falls below 100. Number changed uses the first reading as its baseline. Each region reads its first number, so avoid including unrelated values. Speed and accuracy depend on image clarity, fonts and your device.

Save up to 8 source configurations and connect 4 screens, windows or tabs at once. Each source has independent tasks and monitoring state. Browsers with Document Picture-in-Picture can open a floating window: controls act on the selected source, results and records cover all sources, and tasks are grouped by source. Closing it does not stop sharing. Keep the main page open.

OKYA Watch monitors the visible image you choose to share, including text, numbers or colors in webpages, application windows and videos. It does not fetch webpages on a server schedule from a URL or refresh the target page automatically. Keep the main browser page and target source running. Closing the main page stops sharing and recognition.

The interface and OCR both support English, simplified Chinese, Spanish, Portuguese, Japanese, Korean, German, French, Italian and Dutch. OCR language models are downloaded on demand the first time they are used and cached in this browser; interface and OCR languages can be configured independently.

By default, screenshots are processed in this browser's memory and are not uploaded as recognition records. Tasks and records stay local. External notifications or authorized cloud translation send recognized text to services you configure. Image matching stores only a reduced reference pattern.

Choose a point or region. A point averages nearby 3 × 3 pixels; a region uses RGB tolerance and a changed-pixel percentage. Set a target color with the color picker or HEX. A percentage of 0 allows any changed or matching pixel to trigger.

Yes. QR codes can provide text or links. Linear formats include Code 39, Code 93, Code 128, EAN, ITF, UPC and Codabar. Native browser detection is used first, with automatic local decoding as a fallback; the screen is never uploaded.

Sharing can continue while you use another app, but browsers may reduce background recognition frequency. Minimized target apps may stop rendering or become unavailable to capture, so continued updates are not guaranteed. Share a specific window or tab and keep it running without minimizing it.

Browser-local translation is selected by default and requires a supported browser and language pair. Otherwise, configure a cloud translation endpoint and explicitly authorize sending recognized text. AI visual understanding is not available, and the product does not control other apps' mouse or keyboard.

Yes. Share a playing video window or tab, create a screen text extraction task for the subtitle region, and select Video subtitles mode. It preprocesses the region before local OCR. For translation, use a supported browser-local translator or configure and authorize a cloud endpoint. Results depend on video quality, subtitle style and playback.

Yes. Select in-page, system, Webhook, Telegram and email notifications independently for each task. Webhooks support Slack and Discord. Telegram needs a Bot Token and Chat ID; email needs your own notification endpoint.