Every file has a
job to get done.
This is the bench
you do it on.

Why use Mergio for your PDF, image, video, and audio files?

Most online file converters follow the same model: upload your document to a remote server, wait in a queue while it processes, download the result — often behind a paywall, daily limit, or watermark. Your private contracts, medical records, financial statements, and personal photos pass through someone else's infrastructure, get stored on disks you don't control, and may be retained for days or indefinitely. Mergio takes a fundamentally different approach.

Every one of Mergio's 150+ tools runs entirely inside your browser tab. When you drop a PDF into the Merge tool, the file is read by the browser's native File API, processed in-memory using JavaScript libraries like PDF-lib and PDF.js, and the output is generated as a local Blob URL. At no point does any network request carry your file data to an external server. The same is true whether you're compressing a video (processed via FFmpeg compiled to WebAssembly), removing an image background (processed via Canvas pixel manipulation), or running OCR on a scanned document (processed via Tesseract.js running a trained model locally). This isn't marketing language — you can verify it yourself by opening your browser's Network tab during any operation.

The practical benefit is immediate: no upload wait on slow connections, no risk of a server breach exposing your files, and no artificial limits on how many files you process per day. A freelance graphic designer can batch-compress 50 client photos in a row. A law office can merge and sign sensitive contracts without sending them to a third party. A student can convert lecture recordings between audio formats without creating yet another account on yet another SaaS tool.

Mergio is organized into seven specialized benches, each focused on a file type:

• PDF Bench (32 tools) — merge, split, compress, sign, OCR, form filling, page extraction, metadata editing, format conversion to JPG/PNG/Word/Excel/Markdown/JSON
• Image Bench (22 tools) — compress, resize, crop, convert formats, remove backgrounds, blur, watermark, make collages and memes, extract EXIF data
• Video Bench (23 tools) — compress, trim, convert, extract audio, add text/images/subtitles, record screen, change speed, merge clips
• Audio Bench (17 tools) — trim, change pitch/speed/volume, join tracks, remove vocals, reduce noise, normalize, compress, record
• Text Bench (22 tools) — OCR (image and PDF), convert between TXT/PDF/Word/HTML/Markdown, format JSON/XML, encode/decode Base64 and URLs
• Converters Bench (28 tools) — move files between formats: HEIC↔JPG, WebP↔PNG, SVG↔PNG, GIF↔MP4, MOV/MKV/AVI→MP4, WAV/FLAC/M4A→MP3, DOCX↔TXT, XLSX↔CSV, EPUB→Text
• Utility Bench (10 tools) — generate QR codes and barcodes, create secure passwords, compute file hashes (SHA-256), inspect file metadata, pick and convert colors

Because everything runs locally in your browser, there's no account to create, no software to install, and no watermark stamped on your output. If you want to understand how browser-based file processing works, the MDN File API documentation and the FFmpeg.wasm project page are good starting points.

How it works under the hood

Mergio doesn't use a single processing engine — each bench leverages the browser technology best suited to that file type. Here's a closer look at the technical stack behind each category:

PDF processing: PDF-lib + PDF.js

PDF creation, merging, splitting, metadata editing, form filling, and signing use PDF-lib, an open-source library that manipulates PDF document structures (cross-reference tables, page trees, content streams) directly in memory. PDF rendering — needed for compression, OCR, thumbnail previews, and image export — uses PDF.js (Mozilla's browser-based PDF renderer) to paint each page onto an HTML5 Canvas, which can then be re-encoded at a target quality level or fed into Tesseract for text recognition.

Image processing: Canvas API + ImageData

Every image tool draws the source image onto an off-screen HTML5 Canvas, then reads or writes pixel data through the ImageData interface. Compression works by re-encoding the canvas to JPEG or WebP at a user-specified quality factor. Filters like blur, sharpen, and grayscale apply convolution matrices or per-pixel color math. Background removal uses tolerance-based flood fill on the alpha channel. Format conversion exploits the browser's built-in canvas.toBlob() with target MIME types.

Video and audio processing: FFmpeg WebAssembly

Video and audio tools use FFmpeg compiled to WebAssembly via the @ffmpeg/ffmpeg library. This gives Mergio access to the same codec library used by professional video editors — H.264 encoding via libx264, MP3 encoding via libmp3lame, and dozens of audio/video filters — all running inside a browser Web Worker. The WASM binary (~25 MB) is downloaded once on first use and cached by the service worker. Because FFmpeg runs in a worker thread, the browser UI stays responsive during encoding. Cross-Origin Isolation headers (Cross-Origin-Embedder-Policy and Cross-Origin-Opener-Policy) enable SharedArrayBuffer, which FFmpeg needs for multi-threaded operation.

OCR: Tesseract.js

Tesseract.js runs a trained LSTM neural network model entirely in the browser using WebAssembly. When you use the Image-to-Text or PDF OCR tool, the image or rendered PDF page is passed to Tesseract's recognition engine, which identifies text regions, segments characters, and produces a UTF-8 text output with confidence scores. The English language model (~4 MB) is loaded on demand.

Document and spreadsheet conversion: JSZip + OpenXML

DOCX, XLSX, and PPTX files are ZIP archives containing XML documents. Mergio uses JSZip to unpack these archives in the browser, parse the XML content (paragraphs, cells, shared string tables), and either extract the text or rebuild the archive in a different format. Creating an XLSX from CSV, for example, involves constructing a complete OpenXML workbook package with content types, relationships, and worksheet XML — all generated client-side.

Security tools: Web Crypto API

The Password Generator uses crypto.getRandomValues(), a cryptographically secure pseudorandom number generator (CSPRNG) built into every modern browser. File hashing uses crypto.subtle.digest() to compute SHA-256 and SHA-1 checksums at hardware-accelerated speed. These are the same primitives used by HTTPS and password managers — not a JavaScript Math.random() wrapper.