Image to Text (OCR)

BETA

Extract editable text from an image. OCR runs entirely in your browser with a local Korean PP-OCRv5 model.

Quick answer and specifications

The core workflow runs in this browser. Check the supported formats, published limits, and any on-demand model download before you start.

Inputs
JPG, PNG, WebP, & GIF
Outputs
text & Markdown
Processing
On this device, in the browser
Model languages
Korean and English
Per-file limit
10 MB
Files per run
1
On-demand model download
18 MB
Last verified

The interface language does not determine the model language. Recognition and pronunciation quality vary; check the result.

OCR runs locally in this browser. Your image is never uploaded to Aiviko or an OCR API.

Drop an image here

PNG, JPG, WebP, or GIF · one image · up to 10 MB

Browse files

Extracted text

Choose an image to see its extracted text here.

Continue with a focused workflow

Image to Text (Local OCR) →Screenshot to Text →Korean OCR →

How to extract text from an image locally

Aiviko's Image to Text (OCR) tool extracts editable Korean, English, and numeric text from a PNG, JPG, WebP, or GIF image up to 10 MB (up to 50 megapixels) directly in your browser. It runs a local Korean PP-OCRv5 detection and recognition pair (about 18 MB downloaded on first use and cached) inside a dedicated Web Worker with WebGPU acceleration and WASM fallback.

  1. 1. Drop or paste one image (up to 10 MB)

    Select a clear PNG, JPG, WebP, or GIF photo, scan, or screenshot up to 10 MB and 50 megapixels.

  2. 2. Run local PP-OCRv5 inference

    On first run, the browser downloads and caches about 18 MB of OCR model files, then executes recognition in a background worker via WebGPU or WASM.

  3. 3. Review and export TXT or Markdown

    Compare names, numbers, and line breaks against the original image, then copy the text or download TXT or Markdown.

Supported OCR inputs and workflows

  • Korean, English, and numeric text extraction from receipts, notices, forms, and UI screenshots using a local Korean PP-OCRv5 model
  • Single-image input for PNG, JPG/JPEG, WebP, and GIF files up to 10 MB and 50 megapixels (with automatic downscaling into the 8 MB internal prepared budget)
  • Hardware-accelerated ONNX Runtime Web execution with automatic WebGPU selection and version-pinned WASM fallback
  • One-click clipboard copy plus plain-text (.txt) and Markdown (.md) file downloads

Model scope and accuracy limitations

  • The bundled PP-OCRv5 recognition model is trained for Korean, English, and numbers; it does not provide dedicated Japanese, Chinese, or Spanish character recognition.
  • PDF documents and iPhone HEIC photos are not accepted directly—convert PDF pages or HEIC files to PNG or JPG in the File Converter first.
  • Handwriting, vertical text, complex multi-column tables, formulas, and blurry or tilted photos can produce omissions or misordered lines; always verify critical figures against the source.

Related browser tools

File Converter →

Convert PDF pages or iPhone HEIC photos into PNG or JPG before running OCR.

Background Remover →

Remove backgrounds from portrait photos locally using a 25 MB on-device MODNet model.

Frequently asked questions about Image to Text (OCR)

Can I extract text from a PDF or iPhone HEIC photo?

OCR accepts PNG, JPG, WebP, and GIF images up to 10 MB. If you have a PDF or HEIC file, convert it to PNG or JPG with Aiviko's local File Converter first, then drop the exported image into OCR.

What languages does this OCR model recognize?

Although the Aiviko interface is available in five languages, the current OCR engine uses a Korean PP-OCRv5 recognition model designed for Korean, English, and numeric text.

Are my document photos uploaded to an OCR server?

No. Image preparation and PP-OCRv5 inference run inside a browser Web Worker on your device. The browser only downloads the ~18 MB model and runtime files on demand.

Before you use this tool

Product information reviewed by Aiviko Team

Where processing happens

The selected input is processed in your browser and is not uploaded to an Aiviko processing server.

What may use the network

The page may download code, fonts, runtime files or machine-learning models. Analytics and advertising load only under the choices and conditions described in the privacy and cookie policies.

Results and limits

Speed, accuracy and maximum practical file size depend on the input, browser and device. Review the guide before using the output for an important decision.

Read the full guide and troubleshooting notes →