First recognize the text, then infer the rows and columns based on position and spacing, and export the TSV draft.
Select a text image, up to 12 MiB, 4 Megapixels.
The first time you use it, the recognition program and language data will be loaded from this website, and the pictures will not be uploaded. The maximum image size is 4 million pixels, and the results require manual proofreading.
Infer rows and columns based on text position and spacing, and only generate a TSV draft to be proofread; merged cells are not recognized, and complex tables may be misaligned.
These are demo inputs and sample results, and do not mean that your data has been processed.
It is not a high-precision table structure model; it does not recognize merged cells, and complex tables are easily misaligned. Results must be manually proofread, and formula-style text will have a security prefix added.
Content check:2026-09-08 · About this site and content description