@remotion/media-parser

available from v4.0.190

This is a package that parses video and audio files to extract metadata and samples.

warning

Unstable API: This package is experimental. The API may change in the future.
The API for getting video metadata is stable and may be used in production.

Design goals:

• Support all major containers: .mp4, .mov, .webm, .mkv, .avi, .m3u8, .ts, .mp3, .wav, .aac and .flac.
• Easily get various information from multimedia files
• Work in the browser, and server runtimes (Node.js, Bun, Edge, etc.)
• Satisfy your query with minimal fetching
• Functional TypeScript API
• Be useful when passing unsupported media
• Allow decoding video frames and audio samples using WebCodecs
• Pausable, resumable and cancellable
• Seeking - continue parsing in a different section of the file (not yet implemented)
• No dependencies

Installation

npm

npm i --save-exact @remotion/media-parser@4.0.271

yarn

pnpm i @remotion/media-parser@4.0.271

pnpm

bun i @remotion/media-parser@4.0.271

bun

yarn --exact add @remotion/media-parser@4.0.271

This assumes you are currently using v4.0.271 of Remotion.
Also update remotion and all `@remotion/*` packages to the same version.
Remove all ^ character in front of the version numbers of it as it can lead to a version conflict.

Guide

Getting video metadata

Simple examples of extracting video metadata

Available fields

Information you can get using the media parser

Fast and slow operations

Efficently use parseMedia()

Readers

Read from a variety of sources

Pause, resume and abort

Steer the parsing process

Format support

What you can parse

Runtime support

Where you can run it

Extract ID3 tags and EXIF data

Get embedded tags from video files

Parsing on Web Workers

Parse a media file in the browser on a separate thread.

Download and parse

Download a media file to disk and parse it simultaneously

Foreign file types

Get information from the errors when passing unsupported file types

WebCodecs

Decode video and audio frames in the browser

Stream selection

Choose which streams to use in a HLS Playlist

APIs

The following APIs are available:

parseMedia()

Parse a media file.

downloadAndParseMedia()

Download and parse a media file.

parseMediaOnWebWorker()

Parse a media file in the browser on a separate thread.

parseMediaOnServerWorker()

Parse a media file on the server on a separate thread.

mediaParserController()

Pause, resume and abort the parsing.

hasBeenAborted()

Determine from an error if the parsing has been aborted.

Readers

Choose the appropriate reader for your file source:

nodeReader

Read a file from the local file system.

webReader

Read a file from a File or from a URL.

universalReader

Read a file from a File, from a URL or from the local file system

Writers

Choose how to store files downloaded using downloadAndParseMedia():

nodeWriter

Write a file to the local file system using Node.

How does this compare to FFmpeg?

FFmpeg generally is a more mature software and carries more capabilities.
We advise you to use it always if you don't need to call it from JavaScript.

We built @remotion/media-parser with the focus on making it work well on the web:

• Importing this library is a lot lighter than FFmpeg.wasm, which loads at least 10MB.
• Functional, promise-based API
• Direct integration and intentional usage of JavaScript APIs like fetch(), File, Worker, AbortController, node:fs, resizable ArrayBuffer
• Samples are returned in a WebCodecs-compatible format, which makes it easier to decode them in the browser with full hardware acceleration.

How does this compare to MP4Box.js?

We aim to support more container formats than MP4Box.js, so that multimedia files of any kind can be passed in.
MP4Box.js focuses on the ISO Base Media Format and supports reading more boxes, however it does not support reading other container formats.

Thank you

Tella for boosting the development of @remotion/media-parser with $10.000!

License

Remotion License