randarium
Datasets

Fake Media Metadata Generator

Create synthetic metadata rows for images, videos, and audio files with realistic properties like dimensions, formats, codecs, bitrates, and timestamps. This data is fictional and suitable for testing media processing pipelines.

Also known as: image metadata · video metadata · media file properties

seeded · synthetic data

Presets

Output

No output yet — set your options and hit .
About this tool, tips & examples

What it does

The Fake Media Metadata Generator produces synthetic metadata rows for images, videos, and audio files — dimensions, formats, codecs, bitrates, durations, file sizes, and timestamps — without any actual media behind them. Generate up to 10,000 rows per run, seeded for reproducibility, and export as CSV, JSON, NDJSON, TSV, or Markdown.

Common use cases

  • Media pipeline testing — transcoding queues, thumbnail services, and validation logic exercised against varied codec/format combinations.
  • Asset manager UIs — libraries, grids, and detail panes populated with plausible file properties.
  • Database schema testing — realistic media tables for imports, migrations, and query tuning.
  • Upload validation — rule checks (max resolution, allowed codecs, bitrate ceilings) against data that spans the ranges.

Settings

  • Media Type — image, video, or audio; each type emits its own appropriate properties (dimensions for images, codecs and durations for video/audio). Presets jump to each.
  • How many — 1 to 10,000 rows per run.
  • Seed — identical seed + settings = identical dataset.

Privacy note

Metadata is generated locally in your browser and never uploaded. There are no real files — no EXIF from anyone’s photos, no fingerprints of real media — just fabricated properties in realistic ranges.

FAQ

Are the property combinations realistic? Values come from plausible real-world ranges (common resolutions, codecs, and bitrates), so filters and display logic meet data shaped like production.

Can I test my validation limits? Yes — generate a large batch and the spread of sizes, dimensions, and bitrates will cross typical rule boundaries, exercising both accept and reject paths.

Where are the actual files? There are none by design — this tool tests everything around media (records, rules, UIs) without storing a single byte of it. For fake filenames and paths, see the Fake File generator.