Drop your .csv file here, or click to browse
.csv or .tsv, up to 400 KB
How the CSV translator works
A CSV export mixes text a human should read with data a machine needs untouched: product descriptions next to SKUs, support replies next to ticket IDs, app copy next to translation keys. This tool lets you pick exactly which columns get translated and replaces only those cells inside the original file. The delimiter, the quoting, the line endings, the header row, and every unselected column come back byte-identical, so the file still loads in Excel, Google Sheets, or your import script without a single surprise.
Commas and quotes do not break it
Quoted cells that contain commas, line breaks, or doubled quotes are parsed the way spreadsheets write them, and if a translation itself needs a comma or a quote, the cell is re-quoted correctly on the way out. Semicolon-separated files, the European Excel default, and tab-separated .tsv files are detected automatically.
What is left alone on purpose
- Every column you did not select, byte for byte
- The header row, when you mark the first row as a header
- Numbers, prices, dates, URLs, email addresses, and code-style values, even inside selected columns
- Placeholders like
{name}and%sin app-string exports, masked before the model sees them - The file frame itself: delimiter, quote style, and line endings
Tips for CSV files
- Untick ID and key columns even if they look like words. What ships to production should never be guessed
- Add a glossary for product names so every row uses the same term
- Exports from Excel in Europe often use semicolons. That is fine, the tool detects it
- Spot-check a few rows in a spreadsheet before importing the result anywhere that matters
Frequently asked questions
Can I translate only some columns of my CSV?
Yes, that is the whole point. After you upload the file you get a preview with a checkbox per column. Untick the ones that must stay untouched, like IDs, SKUs, and keys, and only the ticked columns are translated. Every other cell comes back byte-identical.
Will the file still open correctly in Excel or Google Sheets?
Yes. The delimiter, the quoting, the line endings, and the header row are all preserved exactly. If a translated cell needs a comma or a quote of its own, it is re-quoted the way spreadsheets expect, so the file keeps the same number of rows and columns.
Does it handle semicolon-separated files from European Excel?
Yes. Comma, semicolon, tab, and pipe separated files are detected automatically, and the output uses the same separator as the input. Tab-separated .tsv files work too.
What about cells with commas, quotes, or line breaks inside?
Quoted cells that contain commas, doubled quotes, or line breaks are parsed the way spreadsheets write them, translated as one cell, and written back with correct quoting. The row and column structure never changes.
Will it translate my numbers, prices, or product codes by mistake?
No. Even inside a selected column, numbers, prices, dates, URLs, email addresses, and code-style values are skipped automatically. Placeholders like {name} or %s in app-string exports are protected before the AI sees the text, and a cell whose placeholder would not survive is kept in the original language.
Is it free, and are my files stored?
Free, with no signup. Your file is processed in memory to produce the translation and is not kept afterwards, never used for training, and never shared.