You probably got here from an old link about the MP3 format when it was new. Read on a bit about the format, get nastalgic about the first time you heard about this thing called MP3's, and then spend a little quality time exploring the Oddmusic site by clicking on the Oddmusic logo above.

What is MP3?

This page serves as a general reference to the MP3 format and also as a bit of a historical document about the format. Since it has been online since the 1990s, a lot of sites still link to this page from back in the days when the word "MP3" was still a new term. It is still an interesting read, showing how fast things have changes over the last few years, so we'll leave it mostly intact, and add a few new comments in this color.

MP3 is a compression form (like .zip). The name MP3 stands for Mpeg 1 Audio Layer 3 and its compression algorithm is based on a complicated psycho-acoustic model. This model is based on the fact that the human ear cannot hear all the audio frequencies. The human hearing range is between 20Hz to 20Khz and it is most sensitive between 2 to 4 KHz. This MP3 model among other techniques tries to eliminate the frequencies which the human ear is unable to hear keeping all the hearing frequencies leaving intact the hearing experience. As a result the MP3 compression form is destructive. That means that the compression algorithm causes the file to lose some information so that it cannot be restored to its original content.

When compressing an audio file (encoding) using the MP3 algorithm you can set the desired encoding level and have any compression level you want. The larger the compressed file the better the audio quality of the result. The majority of the files available on the internet are encoded in 128 kbits stereo 44khz which results to a high quality file that is 12 times smaller than the original. This of course has many interesting consequences.

1. It's now possible to create CDs that contain over 160 songs and can play for over 14 hours on a personal computer.

2. We can efficiently store songs on a hard disk and play them directly form there. (and now of course, we have standalone MP3 players galore)

3. And best of all we can now easily transmit a song over the internet (a compressed 4 minute song is smaller than 4MB, while uncompressed would be over 40MB!) That's why MP3 has become the hottest keyword on the Internet.

The main disadvantage of the MP3 format is that it needs to be decompressed when played. That means that we need a computer to play them and that we cannot hear them in our ordinary home stereo cd players. This is something that might be changed as many large hardware companies are thinking to support MP3 format in their products. (obviously, now there are standalone players of many types, and most CD players also play MP3s)

How do I play an MP3 file?
You need an MP3 player on your computer. At the time this document was first written in the 1990s, there was some work you had to do, programs to install, etc, but now all modern computer operating systems have built in media players that will play MP3s and many other audio and video formats. Many folks like to use 3rd party players like Winamp or iTunes nowadays, but either way, chances are if you click on an MP3 file on your computer, some program is going to automatically start up and play it.

Why compress audio?
There is a simple reason to compress digital audio. High quality uncompressed digital audio takes a lot of space. Just 1 minute of CD quality uncompressed audio takes around 10Mb and needs around 50 min to be transmitted over a 28.8 modem! (yes, this was back in the day of mostly dial-up internet connections :-)

Digital audio coding tries to compress digital audio so as to take less space. Modern perceptual audio coding techniques like MP3 (MPEG Layer 3), use the human ear perception of sound to achieve compression factors of 12 with no pereptible loss of quality.

The two parts of audio compression
Audio compression consists of two parts.
1. The Encoding which transforms digital audio data into a highly compressed form called bitstream (or coded audio data).
2. The Decoding which reconstructs digital audio data from the bitstream. This is absoletely necessary if you want to play the bitstream on your soundcard.

Audio Quality
Please note that what you end up with after encoding and decoding is not the same sound file anymore. Redundant and irrelevant (for the human auditory system) parts of the sound signal are deleted. The reconstructed audio file differs from the original, but it will sound the same - more or less, depending on how much compression had been performed on it.