Hear the actual difference between any lossless and lossy files (Tutorial + Example)

[Please read this whole description before commenting in order to make sure you haven't misinterpreted my intentions for this video. The purpose of this video is to demonstrate a technique that you can use on your own, it's not meant to compare MP3 to a lossless format.]

This is a method compare the difference between a lossless file and a compressed version of that file. This isn't just limited to MP3s - any audio compression that doesn't speed up, slow down or change the volume in audio can be tested with this technique provided that you have the original, uncompressed version that was used to create the compressed version.

What this does is cancel out everything except for the differences in the two files by inverting one and mixing them after they're both lined up. Anything that's not present or has been added during compression can be heard. I used Adobe Audition to do this, but Audacity works as well.

I used a slightly sped-up version of the song "The Bad Touch" because it was easy to line up the lossless and lossy versions, and it has both high and low frequencies. The lossy version of the file was encoded at 192Kbps using Adobe Audition 3.0.1.

Additionally, I normalized the mixed track in order to allow it to be heard better. But the purpose of this video is a tutorial so you can do this yourself and experiment more, allowing you to hear it at its original volume or whatever your heart desires.

Addendum:

YouTube does use compression so this isn't a fair comparison, but is more intended as a demonstration of a technique you can try on your own. Still, just for fun, I did try comparing the two after YouTube compressed the video's audio.

And you can compare those to how it looks in the video. Obviously the high frequencies are cut off and there is some clipping that was introduced apparently, but there are still noticeable differences between the two. Probably not audible differences, though.

Just in case anyone is wondering, to get the video's audio to make those images, I used a service that downloads the direct audio stream from YouTube to an M4A file (no re-compression, just the straight audio data from YouTube ), then used NeroAAC to decode that into a .wav file. No transcoding used, let alone transcoding to MP3.

[junits]

I think you'e the first person I've ever seen that actually proved there was a difference and displayed it in a way that is easy to understand without throwing around audio nonsense words.

[Quarker]

Not sure if people are forgetting this, but this very YouTube video has undergone compression itself. So people perceiving a difference between the lossy and lossless tracks are under the beautiful phenomenon known as the placebo effect. There is an extremely minimal difference between them in this YT video.

[PeterBondeVillain]

You should work in radio.
I could you listen to your smooth, smooth voice all day long.

[davebryan]

I think lossless formats are like fine bottles of wine, and 320 kbps mp3 files are like fine bottles of wine with one tablespoon of water replacing one tablespoon of wine. The frequencies you're missing out on with 320kbps are so minuscule and barely take away from any of the original quality when they're removed. Unless these audiophiles have canine hearing and $1000+ headphones, I really think they're making a big fuss about nothing. The music is almost always equally enjoyable with either format. 

[SirMo]

While it is true, you are missing all those sounds in the inverted difference. It is important to note that those sounds were purposely left out by the compression algorithm because they are masked by louder sounds in the song and due to this your ear can't really hear them anyways. Which is why compressed audio sounds pretty indistinguishable to the raw files. So while yes in the vacuum it does sound like there is a lot of information missing, when combined with the compressed information the difference becomes near impossible to tell apart. It's like laughing in a movie theater. If you laugh during a loud scene no one really hears your laugh. But if a scene is quiet everyone hears it.

[alskjdfsldkjf]

If the energy used to discuss the completely inaudible difference between a 320 kbps mp3 and a FLAC or AIFF file could be converted to power we could stop drilling for oil tomorrow.

[Leonhart_]

The video is correct in the sense that an mp3 is less audio information in the file, but it's not like they remove it randomly. It's based on the theory on psycho-acoustics that states a formula for which some sounds are masked by others entirely. Those sounds are removed along with some very high inaudible frequency. A high-quality mp3 will not be distinguished from lossless by hearing them side by side.

[uheuheuheheu]

1:24 lossless
1:34 mp3
1:44 difference

[pl]

I've just begun learning what it means to have HiFi audio and this video was a perfect first step! Even with Youtube's compression and my LoFi hearing equipment getting in the way, the difference between FLAC and MP3 that you showed is clear. I think it helps to people new to this like I am to notice that what gets cuts in lossy formats is not so much the up-front melodic elements, but the depth/reverb/sustain/echo of the percussion in the back. For drums in particular, some of that "attack" is gone. Would affect some genres less than others, but for me who listens to metal, it's made a noticeable difference now that I've learned to notice. Thanks a bunch.

EDIT: by 'some of the "attack" is gone, ' I mean the texture of the attack. Lower bitrate snare drum hits a tad flatter, more uniform thud while higher bitrate snare drum hits smoother since there's that little extra depth from less audible frequencies present behind it. Fuller sound.

[ustbot]

LETS DOIT LIKE THEY DO ON DISCOVERY CHANNEL

[david-bsxo]

YOUR VOICE IS SO CLEAR AND BUTTERY SMOOTH

[alberteinsteinthejew]

Technically, both audio tracks (the lossless and mp3) are compressed here, since all YouTube videos and their soundtracks are compressed :D

[trophywolfe]

I did this with a 320kb/s m4a file and a 24 96khz flac file and I was stunned. You can nearly hear the entire song.

[CASSEMELLO]

You should have your own radio station. Or work for Audio books company or something. I can listen to you all day lol

[rimereit]

Thank you for the upload. What this shows me is that you have to use software to be able to actually tell the difference between flac and mp3. Turns out mp3 is not as shitty as I thought.

[azerohiro]

If you have a pair of cheap headphones then it really doesn't matter whether you listen to MP3 or FLAC. On higher end models though, the difference between MP3 and FLAC is actually pretty noticeable. But it's not like the difference is worlds apart and you're not really missing out on much.

[blinkvideo]

I've been trying to find differences between lossy and lossless audio. I have a bit of hearing loss, so it's been a challenge. My first thought was "Dude is posting this on Youtube which means it'll be compressed anyway". Thanks for pointing that out around 1:20. =)

[formlessmist]

YOUR VOICE IS SMOOTHER THAN A BUTTERED DOLPHIN

[rutgermuller]

As much as I like this smart comparison method (it has potential for sure), several conclusions made do not make sense.

1 "Wow, that is all you are missing" is not true, because of psychoacoustics: we don't literally hear what goes into our ears, we hear the sound after our brain has processed it. Compression like MP3 makes use of exactly that fact. It is made to trick our brains in complex ways. Listening to the 'difference track' alone doesn't take that into account. Even though there is a difference in measurement, it does not mean a difference in perception. 

2 I would strongly recommend not to use a 'slighty sped up' version of an audio file as the base of this test. Time stretching audio already adds artefacts (distortions) to the music, which compressions might not be able to handle well. (This is also why it makes sense to DJ only with losless files, so that any can do as less "damage" as possible) And of course… Youtube audio and video are always compressed.

3 The only true way to do an audio comparison test is still a double blind test. Where are the people doing these on Youtube?

[entropy]

This makes a difference with space and imaging though, when you hear a lossless file you get a clearer image and depth etc. It makes the differences stand out more