Image to text (OCR)
Extract text from images with OCR. Results can vary with image quality, font and document layout. It is not a certified transcript and not infallible recognition.
- No account needed
- Temporary result, expires on its own
- Processed on our servers
The text could not be extracted
Sending your file
Processing your file
- Uploading to the server
- Processing your file
- Preparing the download
Your text is ready
Download it now: the link is temporary and expires without you deleting anything.
How it works
- Select the image Choose a JPEG, PNG or WEBP with readable text. One image per job.
- Process the image Submit the job and wait for text recognition to finish.
- Download the text Download the .txt. Review the result: there may be recognition errors.
This tool reads text from a photo or a scan and leaves it in a .txt file. Quality is not a published percentage: it depends on resolution, contrast, orientation and language. The result is not a Word file and not a PDF with an invisible text layer.
JPEG, PNG and WEBP are accepted, one image per job, with the guest size maximum (25 MB). Available languages are Spanish and English (there is no picker on the form). A sharp ~300 DPI scan, upright and high-contrast, usually does better than a dark, sideways or flash-lit photo. There is no multi-page PDF OCR in this tool. An image above the megapixel cap is rejected before processing.
Maximum time is published. A job that overruns time fails. If nothing usable is recognized, the text file says no text was recognized: we do not invent words. We also do not “correct” the result with a language model. What you see is the recognition output, trimmed of spaces, ready to copy or archive.
Privacy: the input image is deleted when processing finishes. The .txt lasts for temporary retention. We do not send the image to cloud OCR. There is no account. Always review the text before using it in a filing: a blurry ID, an invoice with a stamp or a screenshot with columns can come out with errors, mixed lines or swapped characters.
Use cases: turn a clear handwritten note into text, extract a paragraph from a screenshot, or recover text from a simple scan. It does not replace a professional transcript or production OCR with templates. If the image is heavy, compress it first. If you start from a PDF, rasterize it to JPG with that tool’s page limit and then pick the page you want to read.
Frequently asked questions
Is OCR 100% accurate?
No. Quality depends on contrast, font, noise and tilt. We publish a usable result, not a guarantee of accuracy.
Which languages are installed?
Spanish and English. We do not advertise other languages because they are not in the image.
Can I pass a multi-page PDF?
Not in this tool. It accepts one image. If you have a PDF, convert it to JPG page by page first, within that tool’s limit.
What file do I download?
A plain text file (.txt), not a Word file and not a PDF with a text layer.
Is the image sent to a cloud OCR service?
No. Recognition happens on ToolZen. The input is deleted when it finishes; the .txt expires with retention.
Do resolution, contrast, orientation and language matter?
Yes. A sharp, high-contrast, upright photo in Spanish or English usually does better. Sideways, blurry or other-language text lowers quality. There is no language picker: recognition uses Spanish and English together.