"The Effects of MP3 Compression on Perceived Emotional Characteristics in Musical Ins

[SEAWOLF97]

.
"The Effects of MP3 Compression on Perceived Emotional Characteristics in Musical Instruments"

http://www.innerfidelity.com/content...tOL353Pzekk.99

[Ed Zeppeli]

With those bit rates I'd have negative emotions too.

[rdgrimes]

Twenty undergraduate students with normal hearing, who were not trained in music or audio arts but were considered "attentive listeners" were used in the test. Subjects listened to sustained instrument sounds of the: bassoon, clarinet, flute, horn, oboe, saxophone, trumpet, and violin. These sustained tone tracks have been used in other studies making it easy for future researchers to compare studies; most of the sounds came from the McGill University instrument sound collection.
Each track was compressed using the LAME MP3 encoder to: 112Kbps, 56Kbps, and 32Kbps. The original uncompressed track and compressed tracks were presented to listeners in randomized pairs and were reproduced by a Sound Blaster X-Fi Xtreme Audio sound card and Sony MDR-7506 headphones.

What a joke.

[Ed Zeppeli]

It would have been much more interesting to see if there were emotional differences between 320kbps MP3 and flac, for example.

[rdgrimes]

Most people couldn't tell any difference between HQ VBR MP3 and the original. CBR MP3 is crap at any bit rate.

[hsosdrum]

Lossy compression most obviously fails when attempting to simultaneously reproduce different sounds that all have differently-shaped dynamic waves. Using single steady-state sounds as test material may allow for direct comparison with other studies, but it doesn't correlate at all with how the codec behaves when attempting to reproduce actual music. If you remove the dynamic wave shape of a nylon-string acoustic guitar it will sound identical to a clarinet (they contain nearly identical harmonic components); using such sounds to test a listener's emotional reaction to musical sounds is absurd on its face.

If during the development of the MP3 codec Frauenhofer had used the remix of "Tom's Diner" instead of Suzanne Vega's a capella version, the resulting codec might have done a better job with real music. As it stands now, 320kbps MP3s are barely listenable in my car (in the presence of a fair amount of road noise) and lower-bitrate MP3s are unlistenable, period. At home it's strictly 16/44 CDs or 24/48 files from my studio recording system.

[rdgrimes]

As it stands now, 320kbps MP3s are barely listenable in my car (in the presence of a fair amount of road noise) and lower-bitrate MP3s are unlistenable, period..

A VBR MP3 made with max quality settings will average around 240kbps and sound far better than any 320 CBR files. Anyone who's not familiar with VBR - should be. Virtually any current MP3 player can handle it.

[Ed Zeppeli]

Back when I used MP3 it was all variable bit rate. Now that memory is cheaper, bandwidths are wider and lossless compression are available I don't see any need for MP3 at all.

[SEAWOLF97]

I got the LP of "Wish you were Here" and it contained the chit for the digital download.

When I saw the format was MP3, didn't bother to DL. So after a long time I went back and got it , mainly to see how bad it was.

I don't know it's rates, but was blown away at the great sound. Blindfolded, I would have ID'ed it as FLAC.

The playback system has a 24/192 dac and I think that made some effect too. Have burned it to CD and played through the 250's and am still impressed.

[honkytonkwillie]

What a joke.

Yeah, it does look kind of trivial and not very far-reaching. Most of us wouldn't dispute that mp3s at low bitrates are crap.

But this is how science is done. Very little steps, controlled, documented, and presented in a specific format so it can be reviewed, cited, and repeated by others as necessary.

One or more students probably got a Master's Degree out of this.

I think the paper is interesting for a couple of reasons. First, it's exploring how something can change the emotional perception of sound. I see this as not much different to the way we might compare cables or amplifiers.

Second, from a data perspective, it's interesting to see how they compare and quantify results that are so subjective (which Eb from the oboe sounds more Happy?). Doing statistics on numerical measurements and quantities is easy. But self-reported opinions and estimates of emotional responses? Damn, that's hard. Humans are so variable.

[Doctor_Electron]

Interesting. Drew Daniels did not think much of over-compressed MPdreck either.


http://drewdaniels.com/mp3exposed.html

Regards, - D_E