Pulling the Audio Track Out of a Video File
What this tool does
A video file is really two streams wrapped in one container: the moving picture and a separate audio track. This tool ignores the picture entirely and lifts out just the sound, handing it back as a standalone MP3 or WAV. It opens the container — MP4, WebM, MOV, AVI, MKV, or OGG — decodes the embedded audio with the browser's Web Audio API, then re-encodes it two ways: through the lamejs MP3 encoder for a compressed file, or through a direct 16-bit PCM writer for lossless WAV.
Source files up to 500MB are accepted, and you decide the output shape with the format, bitrate, and channel controls before extraction begins.
When to reach for it
The classic case is turning an MP4 lecture, interview, or music video into an MP3 you can drop onto a phone, a podcast feed, or a playlist. It is equally useful for salvaging the audio from a screen recording, pulling a backing track out of a WebM clip, or preparing a clean audio file to send off for transcription.
Because only the audio matters here, you skip video editors entirely — there is no timeline to scrub and no need to render a whole movie just to keep its soundtrack.
A concrete example
Suppose you have a 90-second MP4 talk. Load it, choose MP3 at 192 kbps in stereo, and press Extract Audio; the track is decoded and run through lamejs, returning a file of roughly 2MB. Switch the channel setting to mono and the result nearly halves — plenty for spoken word. Pick WAV instead and you get an uncompressed 16-bit file that is larger but stays bit-for-bit faithful to the source.
Notes and edge cases
Decoding relies on the codecs your browser already understands, so an unusual codec inside an AVI or MKV may refuse to decode even though the container is on the supported list — re-wrapping the file as MP4 first almost always clears it up. Bitrate ranges from 64 to 320 kbps, where a higher number preserves more detail at the cost of size, and a mono track is about half the size of stereo.
One thing to keep in mind: the entire audio track is decoded into memory before encoding, so a long or high-resolution source is memory-heavy and can run slowly, or run out of memory, on devices with little RAM. Trimming to a shorter clip is the simplest workaround.