How ConvertCove works
Short version: your files are converted by code running in your own browser tab. They are never uploaded, stored or seen by us.
Upload-based converters vs. ConvertCove
Most online converters send your file to their servers, convert it there, and let you download the result. That means waiting for uploads, trusting a company with your photos or spreadsheets, and accepting size limits that exist because server time costs money.
ConvertCove flips that around. When you add a file, the page downloads the right converter (JavaScript and, for heavy codecs, WebAssembly), and your device does the work. The file is read from your disk into memory, converted, and handed back to you as a download link that points to memory in your own browser.
What happens, step by step
- The page loads. Only a small controller script (a few kilobytes) is loaded with the page, so it opens quickly even on mobile data.
- You add a file. ConvertCove recognises it by its extension and immediately fetches the matching engine: image, audio/video, data, font or archive. Codecs such as libheif, libavif or FFmpeg (about 10 MB compressed, for audio and video) are fetched only if one of your files needs them.
- You press Convert. The engine decodes the file, applies your options (size, quality, delimiter and so on) and encodes the target format. Archive work runs in background threads so the page stays responsive.
- You download. Results are created as in-memory “blob” URLs. Closing the tab or pressing Convert more releases them. Nothing is written anywhere else.
What we do not collect
We have no upload endpoint, no file storage and no third-party analytics or advertising scripts. Our web server keeps standard access logs (IP address, page requested, browser type) for security and aggregate traffic statistics. Those logs never contain your files or their names. Details are in the privacy policy.
Honest limits
- Memory: everything happens in your device’s RAM. Per-file limits are 200 MB for images, 1 GB for audio, 2 GB for video (with a warning above 500 MB), 150 MB for data, 50 MB for fonts and 1 GB for archives, but older phones may run out of memory earlier.
- Animation: animated GIF and WebP files are converted using the first frame.
- Metadata: image metadata (EXIF, GPS) is not copied to converted images.
- Fonts: outlines are never rewritten, so a CFF font unpacked from WOFF2 stays an OTF.
- Archives: encrypted and multi-part archives are not supported; RAR and 7z can be extracted but not created.
Open-source software we rely on
ConvertCove stands on excellent open-source libraries. Thank you to their authors.
| Used for | Library | Licence |
|---|---|---|
| Audio and video | FFmpeg via ffmpeg.wasm (single-thread core with x264, libvpx, LAME, Opus, Vorbis) | GPL-2.0+ (core), MIT (wrapper) |
| HEIC / HEIF decoding | libheif (via heic-to) | LGPL-3.0 |
| AVIF encoding | libavif + aom (via jSquash) | BSD-2 / Apache-2.0 |
| WebP fallback encoding | libwebp (via jSquash) | BSD-3 |
| TIFF | UTIF.js | MIT |
| GIF palette & encoding | gifenc | MIT |
| ZIP, GZIP, DEFLATE | fflate | MIT |
| RAR and 7z extraction | libarchive via libarchive.js | BSD-2-Clause (libarchive), MIT (libarchive.js) |
| Excel (XLSX/XLS/ODS) | SheetJS Community Edition | Apache-2.0 |
| YAML | js-yaml | MIT |
| XML | fast-xml-parser | MIT |
| Markdown → HTML | marked | MIT |
| HTML → Markdown | Turndown | MIT |
| WOFF2 | Google woff2 (via woff2-encoder) | MIT |
WOFF (version 1), ICO, BMP, TAR and CSV handling are written by us from the published specifications.