audio-metadata | Extracts meta data from various audio containers | Machine Learning library

audio-metadata is a JavaScript library typically used in Artificial Intelligence, Machine Learning, Nodejs applications. audio-metadata has no bugs, it has no vulnerabilities and it has low support. You can install using 'npm i audio-metadata' or download it from GitHub, npm.

The purpose of this library is to be very fast and small. It's suitable for server-side or client-side. Really any platform that supports ArrayBuffer and its ilk (Uint8Array, etc.). I wrote it because the other libraries were large and very robust; I just needed something that could extract the metadata out without requiring 30KB of JavaScript. audio-metadata.min.js comes in at 6.1K/2.1K minified/gzipped. To accomplish the small size and speed, it sacrifices several things. As such, the code is a bit abstruse, in that you'll see some magic numbers, like offset += 94 where it's ignoring a bunch of header data to get to the good stuff. Don't judge me based on this code. It works and it's tested; it's just hard to read. Of course, since this isn't an actual parser, invalid files will also work. This means, for example, you could only read the first couple hundred bytes of an MP3 file and still extract the metadata from it, rather than requiring actual valid MP3 data.