1159b3361Sopenharmony_ci<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" 2159b3361Sopenharmony_ci "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> 3159b3361Sopenharmony_ci<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> 4159b3361Sopenharmony_ci <head> 5159b3361Sopenharmony_ci <meta name="generator" content="PSPad editor, www.pspad.com" /> 6159b3361Sopenharmony_ci <meta http-equiv="content-type" content="text/html; charset=UTF-8" /> 7159b3361Sopenharmony_ci <link rel="stylesheet" type="text/css" href="styles/lame.css" /> 8159b3361Sopenharmony_ci <title>Introduction to encoding</title> 9159b3361Sopenharmony_ci </head> 10159b3361Sopenharmony_ci <body> 11159b3361Sopenharmony_ci<div id="menu"> 12159b3361Sopenharmony_ci <ul> 13159b3361Sopenharmony_ci <li><a href="index.html">Index page</a></li> 14159b3361Sopenharmony_ci <li><a href="about.html">About LAME</a></li> 15159b3361Sopenharmony_ci <li><a style="border: 0" href="introduction.html">Intro to encoding</a> 16159b3361Sopenharmony_ci <div id="submenu"> 17159b3361Sopenharmony_ci <ul> 18159b3361Sopenharmony_ci <li><a href="cbr.html">CBR </a></li> 19159b3361Sopenharmony_ci <li><a href="abr.html">ABR </a></li> 20159b3361Sopenharmony_ci <li><a href="vbr.html">VBR </a></li> 21159b3361Sopenharmony_ci <li><a style="border: 0" href="ms_stereo.html">M/S Stereo</a></li> 22159b3361Sopenharmony_ci </ul> 23159b3361Sopenharmony_ci </div> 24159b3361Sopenharmony_ci </li> 25159b3361Sopenharmony_ci <li><a href="usage.html">Usage of LAME</a></li> 26159b3361Sopenharmony_ci <li><a href="history.html">Version history</a></li> 27159b3361Sopenharmony_ci <li><a href="contributors.html">Coders of LAME</a></li> 28159b3361Sopenharmony_ci <li><a href="contact.html">Contact LAME</a></li> 29159b3361Sopenharmony_ci <li><a style="border: 0" href="links.html">LAME links</a></li> 30159b3361Sopenharmony_ci </ul> 31159b3361Sopenharmony_ci</div> 32159b3361Sopenharmony_ci<div id="container"> 33159b3361Sopenharmony_ci<div id="content"> 34159b3361Sopenharmony_ci 35159b3361Sopenharmony_ci<div align="center"> 36159b3361Sopenharmony_ci <img src="images/logo.gif" width="358" height="231" alt="LAME Official Logo" /> 37159b3361Sopenharmony_ci <h2 class="hilight">Introduction to encoding</h2> 38159b3361Sopenharmony_ci</div> 39159b3361Sopenharmony_ci<h3> Introduction</h3> 40159b3361Sopenharmony_ci<p> 41159b3361Sopenharmony_ci There is a lot of confusion surrounding the terms audio compression<a href="#note1"> 42159b3361Sopenharmony_ci <sup>1</sup></a>, audio encoding, and audio decoding. This section will give you 43159b3361Sopenharmony_ci an overview what audio coding (another one of these terms...) is all about. 44159b3361Sopenharmony_ci</p> 45159b3361Sopenharmony_ci<h3>The purpose of audio compression</h3> 46159b3361Sopenharmony_ci<p>Up to the advent of audio compression, high-quality digital audio data took 47159b3361Sopenharmony_ci a lot of hard disk space to store. Let us go through a short example. 48159b3361Sopenharmony_ci</p> 49159b3361Sopenharmony_ci<p> 50159b3361Sopenharmony_ci You want to sample 1 minute of your favourite song and store it on your harddisk. 51159b3361Sopenharmony_ci Because you want CD quality, you sample at 44.1 kHz, stereo, with 16 bits per sample. 52159b3361Sopenharmony_ci</p> 53159b3361Sopenharmony_ci<p> 54159b3361Sopenharmony_ci 44100 Hz means that you have 44100 values per second coming in from your sound card 55159b3361Sopenharmony_ci (or input file). Multiply that by two because you have two channels. Multiply by 56159b3361Sopenharmony_ci another factor of two because you have two bytes per value (that's what 16 bit 57159b3361Sopenharmony_ci means). The song will take up 44100 <sup>samples</sup>/<sub>s</sub> · 2 channels 58159b3361Sopenharmony_ci · 2 <sup>bytes</sup>/<sub>sample</sub> · 59159b3361Sopenharmony_ci 60 <sup>s</sup>/<sub>min</sub> ~ 10 MBytes of storage space on your harddisk. 60159b3361Sopenharmony_ci</p> 61159b3361Sopenharmony_ci<p> 62159b3361Sopenharmony_ci In order to stream this over internet, a speed of at least 1.41<sup>Mbits</sup>/ 63159b3361Sopenharmony_ci <sub>s</sub> is needed, 64159b3361Sopenharmony_ci which wasn't a common speed at all at the time MP3 was invented. 65159b3361Sopenharmony_ci If you wanted to download that, given an average 56k modem connected at 44k, it 66159b3361Sopenharmony_ci would take 1.41Mbits · 1000 <sup>kbits</sup>/<sub>Mbit</sub> / 44 kbits ~ 32 67159b3361Sopenharmony_ci times as much. 68159b3361Sopenharmony_ci <br/>This means 32 minutes just to download one minute of music! 69159b3361Sopenharmony_ci</p> 70159b3361Sopenharmony_ci<p> 71159b3361Sopenharmony_ci Digital audio coding, which - in this context - is synonymously called digital 72159b3361Sopenharmony_ci audio compression as well, is the art of minimizing storage space (or channel 73159b3361Sopenharmony_ci bandwidth) requirements for audio data. Modern perceptual audio coding techniques 74159b3361Sopenharmony_ci (like MPEG Layer III) exploit the properties of the human ear (the perception of 75159b3361Sopenharmony_ci sound) to achieve a size reduction by a factor of 11 with little or no perceptible 76159b3361Sopenharmony_ci loss of quality. 77159b3361Sopenharmony_ci</p> 78159b3361Sopenharmony_ci<p> 79159b3361Sopenharmony_ci Therefore, such schemes are the key technology for high quality low bit-rate 80159b3361Sopenharmony_ci applications, like soundtracks for CD-ROM games, solid-state sound memories, 81159b3361Sopenharmony_ci Internet audio, digital audio broadcasting systems, and the like. 82159b3361Sopenharmony_ci</p> 83159b3361Sopenharmony_ci<h3>The two parts of audio compression</h3> 84159b3361Sopenharmony_ci<p> 85159b3361Sopenharmony_ci Audio compression really consists of two parts. The first part, called encoding, 86159b3361Sopenharmony_ci transforms the digital audio data that resides, say, in a WAVE file, into a highly 87159b3361Sopenharmony_ci compressed form called bitstream. To play the bitstream on your soundcard, you 88159b3361Sopenharmony_ci need the second part, called decoding. Decoding takes the bitstream and re-expands 89159b3361Sopenharmony_ci it to a WAVE file. 90159b3361Sopenharmony_ci</p> 91159b3361Sopenharmony_ci<p> 92159b3361Sopenharmony_ci The program that effects the first part is called an audio encoder. LAME is such 93159b3361Sopenharmony_ci an encoder . The program that does the second part is called an audio decoder. 94159b3361Sopenharmony_ci Nowadays there are lots of players that decode MP3 95159b3361Sopenharmony_ci</p> 96159b3361Sopenharmony_ci<h3>Compression ratios, bitrate and quality</h3> 97159b3361Sopenharmony_ci<p> 98159b3361Sopenharmony_ci It has not been explicitly mentioned up to now: What you end up with after 99159b3361Sopenharmony_ci encoding and decoding is not the same sound file anymore: All superfluous 100159b3361Sopenharmony_ci information has been squeezed out, so to say. It is not the same file, but it 101159b3361Sopenharmony_ci will sound the same - more or less, depending on how much compression has been 102159b3361Sopenharmony_ci performed on it. 103159b3361Sopenharmony_ci</p> 104159b3361Sopenharmony_ci<p> 105159b3361Sopenharmony_ci Generally speaking, the lower the compression ratio achieved, the better the 106159b3361Sopenharmony_ci sound quality will be in the end - and vice versa.<br/> 107159b3361Sopenharmony_ci Table 1.1 gives you a rough estimate about the quality you can expect. 108159b3361Sopenharmony_ci</p> 109159b3361Sopenharmony_ci<p> 110159b3361Sopenharmony_ci Because compression ratio is a somewhat unwieldy measure, experts use the term 111159b3361Sopenharmony_ci bitrate when speaking of the strength of compression. Bitrate denotes the average 112159b3361Sopenharmony_ci number of bits that one second of audio data will take up in your compressed 113159b3361Sopenharmony_ci bitstream. Usually the units used will be kbps, which is kbits/s, or 1000 bits/s 114159b3361Sopenharmony_ci (not 1024).<br/> 115159b3361Sopenharmony_ci To calculate the number of bytes per second of audio data, simply divide the 116159b3361Sopenharmony_ci number of bits per second by eight. 117159b3361Sopenharmony_ci</p> 118159b3361Sopenharmony_ci<table align="center" cellpadding="5"> 119159b3361Sopenharmony_ci <caption><strong>table 1.1:</strong> bitrate versus sound quality</caption> 120159b3361Sopenharmony_ci <tr> 121159b3361Sopenharmony_ci <th>Bitrate</th> 122159b3361Sopenharmony_ci <th>Bandwidth</th> 123159b3361Sopenharmony_ci <th>Quality comparable to</th> 124159b3361Sopenharmony_ci </tr> 125159b3361Sopenharmony_ci <tr> 126159b3361Sopenharmony_ci <td>16 kbps mono</td> 127159b3361Sopenharmony_ci <td>5.5 khz</td> 128159b3361Sopenharmony_ci <td>above shortwave radio / telephone</td> 129159b3361Sopenharmony_ci </tr> 130159b3361Sopenharmony_ci <tr> 131159b3361Sopenharmony_ci <td>32 kbps mono</td> 132159b3361Sopenharmony_ci <td>8.5 khz</td> 133159b3361Sopenharmony_ci <td>near AM (medium wave) radio</td> 134159b3361Sopenharmony_ci </tr> 135159b3361Sopenharmony_ci <tr> 136159b3361Sopenharmony_ci <td>64kbps mono, 128 kbps stereo</td> 137159b3361Sopenharmony_ci <td>16 khz</td> 138159b3361Sopenharmony_ci <td>FM radio</td> 139159b3361Sopenharmony_ci </tr> 140159b3361Sopenharmony_ci <tr> 141159b3361Sopenharmony_ci <td style="border-bottom:0px">-V 3~-V 0 (160~200 kbps) <br/> 142159b3361Sopenharmony_ci (variable bitrate) 143159b3361Sopenharmony_ci </td> 144159b3361Sopenharmony_ci <td style="border-bottom:0px">18~20 khz</td> 145159b3361Sopenharmony_ci <td style="border-bottom:0px">perceptual transparency versus CD<a href="#transparency"><sup>2</sup> 146159b3361Sopenharmony_ci </a> 147159b3361Sopenharmony_ci </td> 148159b3361Sopenharmony_ci </tr> 149159b3361Sopenharmony_ci</table> 150159b3361Sopenharmony_ci<div id="notes"> 151159b3361Sopenharmony_ci <ol> 152159b3361Sopenharmony_ci <li><a name="note1"></a>Audio compression (also called coding) 153159b3361Sopenharmony_ci means reduce the size (bytes) that the original source requires to be stored. 154159b3361Sopenharmony_ci This is not the same than compressors in DSP (or audio effects). The latter 155159b3361Sopenharmony_ci reduces the dynamic range of the audio so that there is less difference in 156159b3361Sopenharmony_ci perceived loudness between its strong and subtle parts. 157159b3361Sopenharmony_ci </li> 158159b3361Sopenharmony_ci <li><a name="note2"></a>Lossy encoding (as opposed to lossless) cannot guarantee 159159b3361Sopenharmony_ci transparency all of the time. This is the value accepted as the <i>sweet spot</i>. 160159b3361Sopenharmony_ci </li> 161159b3361Sopenharmony_ci </ol> 162159b3361Sopenharmony_ci</div> 163159b3361Sopenharmony_ci</div> 164159b3361Sopenharmony_ci<div id="footer"> 165159b3361Sopenharmony_ci 166159b3361Sopenharmony_ci<a href="http://sourceforge.net/projects/lame"><img src="images/sflogo.gif" 167159b3361Sopenharmony_ci alt="Get LAME (Lame Aint an MP3 Encoder) at SourceForge.net. Fast, secure and Free Open Source software downloads" 168159b3361Sopenharmony_ci border="0" height="30" width="120" /></a> 169159b3361Sopenharmony_ci 170159b3361Sopenharmony_ci<a href="http://validator.w3.org/#validate_by_upload"><img src="images/valid-xhtml10.png" 171159b3361Sopenharmony_ci alt="Valid XHTML 1.0 Transitional" border="0" height="31" width="88" /></a> 172159b3361Sopenharmony_ci 173159b3361Sopenharmony_ci<a href="http://jigsaw.w3.org/css-validator/"><img src="images/valid-css.png" 174159b3361Sopenharmony_ci alt="Valid CSS!" border="0" height="31" width="88" /></a> 175159b3361Sopenharmony_ci 176159b3361Sopenharmony_ci<p>HTML markup and design by <a 177159b3361Sopenharmony_cihref="http://www.rarewares.org" target="_blank">Roberto Amorim</a> and <a 178159b3361Sopenharmony_cihref="http://www.maresweb.de" target="_blank">Sebastian Mares</a>. Logo by <a 179159b3361Sopenharmony_cihref="http://www.brightercreative.co.uk">Sam Fisher</a>.</p> 180159b3361Sopenharmony_ci 181159b3361Sopenharmony_ci</div> 182159b3361Sopenharmony_ci</div> 183159b3361Sopenharmony_ci 184159b3361Sopenharmony_ci </body> 185159b3361Sopenharmony_ci</html> 186
