1 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
|
---|
2 | <HTML>
|
---|
3 | <HEAD>
|
---|
4 | <TITLE>Introduction</TITLE>
|
---|
5 | <META NAME="description" CONTENT="Introduction">
|
---|
6 | <META NAME="keywords" CONTENT="lame">
|
---|
7 | <META NAME="resource-type" CONTENT="document">
|
---|
8 | <META NAME="distribution" CONTENT="global">
|
---|
9 | <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso_8859_1">
|
---|
10 | <LINK REL="STYLESHEET" HREF="lame.css">
|
---|
11 | </HEAD>
|
---|
12 | <BODY TEXT=#000000
|
---|
13 | BGCOLOR=#F9FBFB LINK=#006666 VLINK=#4C4C4C
|
---|
14 | ALINK=#995500>
|
---|
15 | <FONT FACE = Helvetica ><strong>Subsections</strong>
|
---|
16 | <UL>
|
---|
17 | <LI><A
|
---|
18 | HREF="node6.html#SECTION00311000000000000000"> The purpose of audio compression</A>
|
---|
19 | <LI><A
|
---|
20 | HREF="node6.html#SECTION00312000000000000000"> The two parts of audio compression</A>
|
---|
21 | <LI><A
|
---|
22 | HREF="node6.html#SECTION00313000000000000000"> Compression ratios, bitrate and
|
---|
23 | quality</A>
|
---|
24 | </UL>
|
---|
25 |
|
---|
26 | <HR>
|
---|
27 | <H1><A NAME="SECTION00310000000000000000"> Introduction</A> </H1>
|
---|
28 | There is a lot of confusion surrounding the terms <I>audio compression</I>, <I>audio
|
---|
29 | encoding</I>, and <I>audio decoding</I>. This section will give you an overview
|
---|
30 | what audio coding (another one of these terms...) is all about.
|
---|
31 | <P>
|
---|
32 | <H2><A NAME="SECTION00311000000000000000"> The purpose of audio compression</A>
|
---|
33 | </H2>
|
---|
34 | <P> Up to the advent of audio compression, high-quality digital audio data took
|
---|
35 | a lot of hard disk space to store. Let us go through a short example.
|
---|
36 | <P> You want to, say, sample your favorite 1-minute song and store it on your
|
---|
37 | harddisk. Because you want CD quality, you sample at 44.1 kHz, stereo,
|
---|
38 | with 16 bits per sample.
|
---|
39 | <P> 44100 Hz means that you have 44100 values per second coming in from your sound
|
---|
40 | card (or input file). Multiply that by two because you have two channels. Multiply
|
---|
41 | by another factor of two because you have two bytes per value (that's what 16
|
---|
42 | bit means). The song will take up <NOBR>44100 samples/s · 2 channels
|
---|
43 | · 2 bytes/sample · 60 s/min ~ 10 MBytes</NOBR> of storage space
|
---|
44 | on your harddisk.
|
---|
45 | <P> If you wanted to download that over the internet, given an average 56k modem
|
---|
46 | connected at 44k (which is a typical case), it would take you (at least) <NOBR>10000000
|
---|
47 | bytes · 8 bits/byte / (44000 bits/s) · / (60 s/min) ~ 30 minutes</NOBR>
|
---|
48 | <DIV ALIGN="CENTER"> <B>Just to download one minute of music!</B> </DIV>
|
---|
49 | <P> Digital audio coding, which - in this context - is synonymously called digital
|
---|
50 | audio compression as well, is the art of minimizing storage space (or channel
|
---|
51 | bandwidth) requirements for audio data. Modern perceptual audio coding techniques
|
---|
52 | (like MPEG Layer III) exploit the properties of the human ear (the perception
|
---|
53 | of sound) to achieve a size reduction by a factor of 11 with little or no perceptible
|
---|
54 | loss of quality.
|
---|
55 | <P> Therefore, such schemes are the key technology for high quality low bit-rate
|
---|
56 | applications, like soundtracks for CD-ROM games, solid-state sound memories,
|
---|
57 | Internet audio, digital audio broadcasting systems, and the like.
|
---|
58 | <P>
|
---|
59 | <H2><A NAME="SECTION00312000000000000000"> The two parts of audio compression</A>
|
---|
60 | </H2>
|
---|
61 | <P> Audio compression really consists of two parts. The first part, called <I>encoding</I>,
|
---|
62 | transforms the digital audio data that resides, say, in a WAVE file, into a
|
---|
63 | highly compressed form called <I>bitstream</I>. To play the bitstream on your
|
---|
64 | soundcard, you need the second part, called <I>decoding</I>. Decoding takes
|
---|
65 | the bitstream and re-expands it to a WAVE file.
|
---|
66 | <P> The program that effects the first part is called an audio <I>encoder</I>.
|
---|
67 | <i>LAME</i> is such an encoder . The program that does the second part is called
|
---|
68 | an audio <I>decoder</I>. One well-known MPEG Layer III decoder is <tt>Xmms</tt>,
|
---|
69 | another <TT>mpg123</TT>. Both can be found on <A NAME="tex2html1"
|
---|
70 | HREF="http://www.mp3-tech.org">ww.mp3-tech.org</A> .
|
---|
71 | <H2><A NAME="SECTION00313000000000000000"> Compression ratios, bitrate and quality</A>
|
---|
72 | </H2>
|
---|
73 | <P> It has not been explicitly mentioned up to now: What you end up with after
|
---|
74 | encoding and decoding is not the same sound file anymore: All superflous information
|
---|
75 | has been squeezed out, so to say. It is not the same <I>file</I>, but it will
|
---|
76 | <I>sound</I> the same - more or less, depending on how much compression had
|
---|
77 | been performed on it.
|
---|
78 | <P> Generally speaking, the lower the compression ratio achieved, the better the
|
---|
79 | sound quality will be in the end - and <I>vice versa</I>. Table <A HREF="node6.html#table-soundq">1.1</A>
|
---|
80 | gives you an overview about quality achievable.
|
---|
81 | <P> Because compression ratio is a somewhat unwieldy measure, experts use the
|
---|
82 | term <I>bitrate</I> when speaking of the strength of compression. Bitrate denotes
|
---|
83 | the average number of bits that one second of audio data will take up in your
|
---|
84 | compressed bitstream. Usually the units used will be kbps, which is <SUP>kbits</SUP>/<SUB>s</SUB>,
|
---|
85 | or 1000 <SUP>bits</SUP>/<SUB>s</SUB>. To calculate the number of bytes
|
---|
86 | per second of audio data, simply divide the number of bits per second by eight.
|
---|
87 | <P> <BR>
|
---|
88 | <DIV ALIGN="CENTER"><A NAME="table-soundq"> </A> <A NAME="95"> </A>
|
---|
89 | <TABLE CELLPADDING=3 BORDER="1" width="512" height="225">
|
---|
90 | <CAPTION><STRONG>Table 1.1:</STRONG> Bitrate versus sound quality</CAPTION>
|
---|
91 | <TR VALIGN="TOP">
|
---|
92 | <TD ALIGN="RIGHT" nowrap width="115">Bitrate</TD>
|
---|
93 | <TD ALIGN="RIGHT" nowrap width="67">Bandwidth</TD>
|
---|
94 | <TD ALIGN="LEFT" nowrap width="246">Quality comparable to or better than</TD>
|
---|
95 | </TR>
|
---|
96 | <TR VALIGN="TOP">
|
---|
97 | <TD ALIGN="RIGHT" nowrap width="115">16 kbps</TD>
|
---|
98 | <TD ALIGN="RIGHT" nowrap width="67">4.5 kHz</TD>
|
---|
99 | <TD ALIGN="LEFT" nowrap width="246">shortwave radio</TD>
|
---|
100 | </TR>
|
---|
101 | <TR VALIGN="TOP">
|
---|
102 | <TD ALIGN="RIGHT" nowrap width="115">32 kbps</TD>
|
---|
103 | <TD ALIGN="RIGHT" nowrap width="67">7.5 kHz</TD>
|
---|
104 | <TD ALIGN="LEFT" nowrap width="246">AM radio</TD>
|
---|
105 | </TR>
|
---|
106 | <TR VALIGN="TOP">
|
---|
107 | <TD ALIGN="RIGHT" nowrap width="115">96 kbps</TD>
|
---|
108 | <TD ALIGN="RIGHT" nowrap width="67">11 kHz</TD>
|
---|
109 | <TD ALIGN="LEFT" nowrap width="246">FM radio</TD>
|
---|
110 | </TR>
|
---|
111 | <tr valign="TOP">
|
---|
112 | <td align="RIGHT" nowrap width="115">128 kbps</td>
|
---|
113 | <td align="RIGHT" nowrap width="67">16 kHz</td>
|
---|
114 | <td align="LEFT" nowrap width="246">near CD</td>
|
---|
115 | </tr>
|
---|
116 | <tr valign="TOP">
|
---|
117 | <td align="RIGHT" nowrap width="115">160-180 kbps <br>
|
---|
118 | (variable bitrate)</td>
|
---|
119 | <td align="RIGHT" nowrap width="67">20 kHz</td>
|
---|
120 | <td align="LEFT" nowrap width="246">perceptual transparency</td>
|
---|
121 | </tr>
|
---|
122 | <TR VALIGN="TOP">
|
---|
123 | <TD ALIGN="RIGHT" nowrap width="115">256 kbps</TD>
|
---|
124 | <TD ALIGN="RIGHT" nowrap width="67">22 kHz</TD>
|
---|
125 | <TD ALIGN="LEFT" nowrap width="246">studio</TD>
|
---|
126 | </TR>
|
---|
127 | </TABLE>
|
---|
128 | </DIV>
|
---|
129 | <BR>
|
---|
130 | </font>
|
---|
131 | </BODY>
|
---|
132 | </HTML>
|
---|