---
title: "Lossy Audio"
date: 2023-10-22T23:23:27Z
draft: true
description: "Testing whether I can actually hear the difference between lossy and lossless audio"
type: "post"
tags: ["audio", "experiment"]
---


I have a pair of Bose headphones and I hate them, and I don't have any shortage of reasons why. I have to charge the headphones before using them, they don't support USB-C audio, I can't use the headphones and charge them at the same time, and the audio quality sounds even worse than my lowly earbuds. Or maybe this is just because I'm accustomed to the audio quality of the earbuds so anything else sounds worse? The only use I've found for the headphones is using them on planes since they can do (somewhat underwhelming) noise cancelation.

Actually, I did find one other use for them, which is for [audio fidelity ABX tests](https://abx.digitalfeed.net/list.html). Basically it's a double-blind test, you get three samples, A, B, and X, and you have to identify whether X is A or B.

Unfortunately, either my ears are terrible, or I have yet another reason to hate on these headphones, because I failed the lossless versus 320 kbps lossy audio versus lossless FLACs, getting only 52% correct. In fact, I was even struggling with the 96 kbps test, although I quit halfway since I have more important things to do.

Out of curiousity, I tried transcoding FLACs to even lower bitrates like 48 kbps or 32 kbps, and at that point I could noticeably tell the difference. I could even *see* the difference using [my favorite mpv visualizer script](https://git.exozy.me/a/dotfiles/src/branch/main/mpv/scripts/visualizer.lua). The low bitrate files look a lot more pixelated and the highest frequencies don't appear at all. So I guess that just shows my eyes are better than my ears, or at least my ears plus my hated headphones.