Listening for Coding (.mp3, AAC) Artifacts.

When you hear audio engineers talk about coded audio...

Inevitably you’ll hear conversations centering around “coding artifacts” in one way or another.

Before going any further...If you are landing directly here from somewhere else, I suggest you check out the first article on this topic before continuing!

What are “coding artifacts”, and what exactly is it that we are hearing? Listening for artifacts involves knowing where to listen. In this case, one must listen in the background. It is also easier to hear artifacts if you have a non-coded version of the audio to compare with.

Why is this an issue? It helps to know what happens to audio when it is converted to a (permanently) "lossy" data compression format. What is a "lossy" compression format?

When, for an example, a 128 kilobit mp3 file is created, over 90% of the original audio data is PERMANENTLY thrown away (permanently lost) to make the audio file smaller. This is useful (and vital) for shipping audio across the internet (such as for a podcast), but should NEVER be used as high quality source material if you can help it. Always save your files as .WAV, or .AIFF files when producing your content. Convert to MP3, MP4, or AAC ONLY when producing the FINAL package to be sent to your audience.

HINT: Saving a .MP3 file as a .wav does not restore the original quality. The audio is permanently altered once it becomes a lossy audio file.

Let’s turn to our audio examples to demonstrate what happens when a file is converted to a "lossy" format..

This file has no coding (Linear). It is the audio as produced in our studio.

Here is the Coded (MP3) version for general audience consumption.

Since this audio has passed only ONCE through the MP3 process, it SOUNDS reasonably close to the source. But it is permanently altered.

To hear how much it has been altered, we can take the mp3 file, and subtract its audio from the original (linear) audio file. What is left over are the distortion artifacts that are introduced by the process of becoming an MP3 (bit-rate reduced) file.

Since coding artifacts have no relationship with anything in the recording mix (those artifacts weren’t originally part of the source), subtracting the original mix from the mp3 version reveals these artifacts “as plain as day” for all to hear.

Here is the original audio "subtracted" from the .mp3 file.

Doing this reveals the difference between the two...In this example, the difference shows the coding artifacts (distortion) added by the digital data reduction process.

Knowing what to listen for, listening to the background of our first two clips, you will eventually start to hear this distortion. (Much easier to spot while listening on good headphones).

Now that you are aware of what happens (permanently) to your audio when converting to a compressed (like .mp3) format, you can now see why you will want to preserve your original source files as linear (.WAV, AIFF) files for future use.

If you were to take this bit reduced audio file, and convert it AGAIN to an .MP3 the distortion will add, causing the audio to sound even worse. The more times you pass audio through this process, the more distortion you create.

The takeaway?

ALWAYS Keep your archived source material of your shows as pristine, and as close their original forms as possible!!

Happy producing!