| 1 | <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> |
|---|
| 2 | <html> |
|---|
| 3 | <head> |
|---|
| 4 | |
|---|
| 5 | <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-15"/> |
|---|
| 6 | <title>Ogg Vorbis Documentation</title> |
|---|
| 7 | |
|---|
| 8 | <style type="text/css"> |
|---|
| 9 | body { |
|---|
| 10 | margin: 0 18px 0 18px; |
|---|
| 11 | padding-bottom: 30px; |
|---|
| 12 | font-family: Verdana, Arial, Helvetica, sans-serif; |
|---|
| 13 | color: #333333; |
|---|
| 14 | font-size: .8em; |
|---|
| 15 | } |
|---|
| 16 | |
|---|
| 17 | a { |
|---|
| 18 | color: #3366cc; |
|---|
| 19 | } |
|---|
| 20 | |
|---|
| 21 | img { |
|---|
| 22 | border: 0; |
|---|
| 23 | } |
|---|
| 24 | |
|---|
| 25 | #xiphlogo { |
|---|
| 26 | margin: 30px 0 16px 0; |
|---|
| 27 | } |
|---|
| 28 | |
|---|
| 29 | #content p { |
|---|
| 30 | line-height: 1.4; |
|---|
| 31 | } |
|---|
| 32 | |
|---|
| 33 | h1, h1 a, h2, h2 a, h3, h3 a { |
|---|
| 34 | font-weight: bold; |
|---|
| 35 | color: #ff9900; |
|---|
| 36 | margin: 1.3em 0 8px 0; |
|---|
| 37 | } |
|---|
| 38 | |
|---|
| 39 | h1 { |
|---|
| 40 | font-size: 1.3em; |
|---|
| 41 | } |
|---|
| 42 | |
|---|
| 43 | h2 { |
|---|
| 44 | font-size: 1.2em; |
|---|
| 45 | } |
|---|
| 46 | |
|---|
| 47 | h3 { |
|---|
| 48 | font-size: 1.1em; |
|---|
| 49 | } |
|---|
| 50 | |
|---|
| 51 | li { |
|---|
| 52 | line-height: 1.4; |
|---|
| 53 | } |
|---|
| 54 | |
|---|
| 55 | #copyright { |
|---|
| 56 | margin-top: 30px; |
|---|
| 57 | line-height: 1.5em; |
|---|
| 58 | text-align: center; |
|---|
| 59 | font-size: .8em; |
|---|
| 60 | color: #888888; |
|---|
| 61 | clear: both; |
|---|
| 62 | } |
|---|
| 63 | </style> |
|---|
| 64 | |
|---|
| 65 | </head> |
|---|
| 66 | |
|---|
| 67 | <body> |
|---|
| 68 | |
|---|
| 69 | <div id="xiphlogo"> |
|---|
| 70 | <a href="http://www.xiph.org/"><img src="fish_xiph_org.png" alt="Fish Logo and Xiph.org"/></a> |
|---|
| 71 | </div> |
|---|
| 72 | |
|---|
| 73 | <h1>Ogg Vorbis encoding format documentation</h1> |
|---|
| 74 | |
|---|
| 75 | <p><img src="wait.png" alt="wait"/>As of writing, not all the below document |
|---|
| 76 | links are live. They will be populated as we complete the documents.</p> |
|---|
| 77 | |
|---|
| 78 | <h2>Documents</h2> |
|---|
| 79 | |
|---|
| 80 | <ul> |
|---|
| 81 | <li><a href="packet.html">Vorbis packet structure</a></li> |
|---|
| 82 | <li><a href="envelope.html">Temporal envelope shaping and blocksize</a></li> |
|---|
| 83 | <li><a href="mdct.html">Time domain segmentation and MDCT transform</a></li> |
|---|
| 84 | <li><a href="resolution.html">The resolution floor</a></li> |
|---|
| 85 | <li><a href="residuals.html">MDCT-domain fine structure</a></li> |
|---|
| 86 | </ul> |
|---|
| 87 | |
|---|
| 88 | <ul> |
|---|
| 89 | <li><a href="probmodel.html">The Vorbis probability model</a></li> |
|---|
| 90 | <li><a href="bitpack.html">The Vorbis bitpacker</a></li> |
|---|
| 91 | </ul> |
|---|
| 92 | |
|---|
| 93 | <ul> |
|---|
| 94 | <li><a href="oggstream.html">Ogg bitstream overview</a></li> |
|---|
| 95 | <li><a href="framing.html">Ogg logical bitstream and framing spec</a></li> |
|---|
| 96 | <li><a href="vorbis-stream.html">Vorbis packet->Ogg bitstream mapping</a></li> |
|---|
| 97 | </ul> |
|---|
| 98 | |
|---|
| 99 | <ul> |
|---|
| 100 | <li><a href="programming.html">Programming with libvorbis</a></li> |
|---|
| 101 | </ul> |
|---|
| 102 | |
|---|
| 103 | <h2>Description</h2> |
|---|
| 104 | |
|---|
| 105 | <p>Ogg Vorbis is a general purpose compressed audio format |
|---|
| 106 | for high quality (44.1-48.0kHz, 16+ bit, polyphonic) audio and music |
|---|
| 107 | at moderate fixed and variable bitrates (40-80 kb/s/channel). This |
|---|
| 108 | places Vorbis in the same class as audio representations including |
|---|
| 109 | MPEG-1 audio layer 3, MPEG-4 audio (AAC and TwinVQ), and PAC.</p> |
|---|
| 110 | |
|---|
| 111 | <p>Vorbis is the first of a planned family of Ogg multimedia coding |
|---|
| 112 | formats being developed as part of the Xiph.org Foundation's Ogg multimedia |
|---|
| 113 | project. See <a href="http://www.xiph.org/">http://www.xiph.org/</a> |
|---|
| 114 | for more information.</p> |
|---|
| 115 | |
|---|
| 116 | <h2>Vorbis technical documents</h2> |
|---|
| 117 | |
|---|
| 118 | <p>A Vorbis encoder takes in overlapping (but contiguous) short-time |
|---|
| 119 | segments of audio data. The encoder analyzes the content of the audio |
|---|
| 120 | to determine an optimal compact representation; this phase of encoding |
|---|
| 121 | is known as <em>analysis</em>. For each short-time block of sound, |
|---|
| 122 | the encoder then packs an efficient representation of the signal, as |
|---|
| 123 | determined by analysis, into a raw packet much smaller than the size |
|---|
| 124 | required by the original signal; this phase is <em>coding</em>. |
|---|
| 125 | Lastly, in a streaming environment, the raw packets are then |
|---|
| 126 | structured into a continuous stream of octets; this last phase is |
|---|
| 127 | <em>streaming</em>. Note that the stream of octets is referred to both |
|---|
| 128 | as a 'byte-' and 'bit-'stream; the latter usage is acceptible as the |
|---|
| 129 | stream of octets is a physical representation of a true logical |
|---|
| 130 | bit-by-bit stream.</p> |
|---|
| 131 | |
|---|
| 132 | <p>A Vorbis decoder performs a mirror image process of extracting the |
|---|
| 133 | original sequence of raw packets from an Ogg stream (<em>stream |
|---|
| 134 | decomposition</em>), reconstructing the signal representation from the |
|---|
| 135 | raw data in the packet (<em>decoding</em>) and them reconstituting an |
|---|
| 136 | audio signal from the decoded representation (<em>synthesis</em>).</p> |
|---|
| 137 | |
|---|
| 138 | <p>The <a href="programming.html">Programming with libvorbis</a> |
|---|
| 139 | documents discuss use of the reference Vorbis codec library |
|---|
| 140 | (libvorbis) produced by the Xiph.org Foundation.</p> |
|---|
| 141 | |
|---|
| 142 | <p>The data representations and algorithms necessary at each step to |
|---|
| 143 | encode and decode Ogg Vorbis bitstreams are described by the below |
|---|
| 144 | documents in sufficient detail to construct a complete Vorbis codec. |
|---|
| 145 | Note that at the time of writing, Vorbis is still in a 'Request For |
|---|
| 146 | Comments' stage of development; despite being in advanced stages of |
|---|
| 147 | development, input from the multimedia community is welcome.</p> |
|---|
| 148 | |
|---|
| 149 | <h3>Vorbis analysis and synthesis</h3> |
|---|
| 150 | |
|---|
| 151 | <p>Analysis begins by seperating an input audio stream into individual, |
|---|
| 152 | overlapping short-time segments of audio data. These segments are |
|---|
| 153 | then transformed into an alternate representation, seeking to |
|---|
| 154 | represent the original signal in a more efficient form that codes into |
|---|
| 155 | a smaller number of bytes. The analysis and transformation stage is |
|---|
| 156 | the most complex element of producing a Vorbis bitstream.</p> |
|---|
| 157 | |
|---|
| 158 | <p>The corresponding synthesis step in the decoder is simpler; there is |
|---|
| 159 | no analysis to perform, merely a mechanical, deterministic |
|---|
| 160 | reconstruction of the original audio data from the transform-domain |
|---|
| 161 | representation.</p> |
|---|
| 162 | |
|---|
| 163 | <ul> |
|---|
| 164 | <li><a href="packet.html">Vorbis packet structure</a>: |
|---|
| 165 | Describes the basic analysis components necessary to produce Vorbis |
|---|
| 166 | packets and the structure of the packet itself.</li> |
|---|
| 167 | <li><a href="envelope.html">Temporal envelope shaping and blocksize</a>: |
|---|
| 168 | Use of temporal envelope shaping and variable blocksize to minimize |
|---|
| 169 | time-domain energy leakage during wide dynamic range and spectral energy |
|---|
| 170 | swings. Also discusses time-related principles of psychoacoustics.</li> |
|---|
| 171 | <li><a href="mdct.html">Time domain segmentation and MDCT transform</a>: |
|---|
| 172 | Division of time domain data into individual overlapped, windowed |
|---|
| 173 | short-time vectors and transformation using the MDCT</li> |
|---|
| 174 | <li><a href="resolution.html">The resolution floor</a>: Use of frequency |
|---|
| 175 | doamin psychoacoustics, and the MDCT-domain noise, masking and resolution |
|---|
| 176 | floors</li> |
|---|
| 177 | <li><a href="residuals.html">MDCT-domain fine structure</a>: Production, |
|---|
| 178 | quantization and massaging of MDCT-spectrum fine structure</li> |
|---|
| 179 | </ul> |
|---|
| 180 | |
|---|
| 181 | <h3>Vorbis coding and decoding</h3> |
|---|
| 182 | |
|---|
| 183 | <p>Coding and decoding converts the transform-domain representation of |
|---|
| 184 | the original audio produced by analysis to and from a bitwise packed |
|---|
| 185 | raw data packet. Coding and decoding consist of two logically |
|---|
| 186 | orthogonal concepts, <em>back-end coding</em> and <em>bitpacking</em>.</p> |
|---|
| 187 | |
|---|
| 188 | <p><em>Back-end coding</em> uses a probability model to represent the raw numbers |
|---|
| 189 | of the audio representation in as few physical bits as possible; |
|---|
| 190 | familiar examples of back-end coding include Huffman coding and Vector |
|---|
| 191 | Quantization.</p> |
|---|
| 192 | |
|---|
| 193 | <p><em>Bitpacking</em> arranges the variable sized words of the back-end |
|---|
| 194 | coding into a vector of octets without wasting space. The octets |
|---|
| 195 | produced by coding a single short-time audio segment is one raw Vorbis |
|---|
| 196 | packet.</p> |
|---|
| 197 | |
|---|
| 198 | <ul> |
|---|
| 199 | <li><a href="probmodel.html">The Vorbis probability model</a></li> |
|---|
| 200 | <li><a href="bitpack.html">The Vorbis bitpacker</a>: Arrangement of |
|---|
| 201 | variable bit-length words into an octet-aligned packet.</li> |
|---|
| 202 | </ul> |
|---|
| 203 | |
|---|
| 204 | <h3>Vorbis streaming and stream decomposition</h3> |
|---|
| 205 | |
|---|
| 206 | <p>Vorbis packets contain the raw, bitwise-compressed representation of a |
|---|
| 207 | snippet of audio. These packets contain no structure and cannot be |
|---|
| 208 | strung together directly into a stream; for streamed transmission and |
|---|
| 209 | storage, Vorbis packets are encoded into an Ogg bitstream.</p> |
|---|
| 210 | |
|---|
| 211 | <ul> |
|---|
| 212 | <li><a href="oggstream.html">Ogg bitstream overview</a>: High-level |
|---|
| 213 | description of Ogg logical bitstreams, how logical bitstreams |
|---|
| 214 | (of mixed media types) can be combined into physical bitstreams, and |
|---|
| 215 | restrictions on logical-to-physical mapping. Note that this document is |
|---|
| 216 | not specific only to Ogg Vorbis.</li> |
|---|
| 217 | <li><a href="framing.html">Ogg logical bitstream and framing |
|---|
| 218 | spec</a>: Low level, complete specification of Ogg logical |
|---|
| 219 | bitstream pages. Note that this document is not specific only to Ogg |
|---|
| 220 | Vorbis.</li> |
|---|
| 221 | <li><a href="vorbis-stream.html">Vorbis bitstream mapping</a>: |
|---|
| 222 | Specifically describes mapping Vorbis data into an |
|---|
| 223 | Ogg physical bitstream.</li> |
|---|
| 224 | </ul> |
|---|
| 225 | |
|---|
| 226 | <div id="copyright"> |
|---|
| 227 | The Xiph Fish Logo is a |
|---|
| 228 | trademark (™) of Xiph.Org.<br/> |
|---|
| 229 | |
|---|
| 230 | These pages © 1994 - 2005 Xiph.Org. All rights reserved. |
|---|
| 231 | </div> |
|---|
| 232 | |
|---|
| 233 | </body> |
|---|
| 234 | </html> |
|---|
