cog/Frameworks/Vorbis/doc/libvorbis/overview.html

137 lines
5.8 KiB
HTML

<html>
<head>
<title>libvorbis - API Overview</title>
<link rel=stylesheet href="style.css" type="text/css">
</head>
<body bgcolor=white text=black link="#5555ff" alink="#5555ff" vlink="#5555ff">
<table border=0 width=100%>
<tr>
<td><p class=tiny>libvorbis documentation</p></td>
<td align=right><p class=tiny>libvorbis version 1.3.2 - 20101101</p></td>
</tr>
</table>
<h1>Libvorbis API Overview</h1>
<p>Libvorbis is the reference implementation of the Vorbis codec. It is
the lowest-level interface to the Vorbis encoder and decoder, working
with packets directly.</p>
<p>All libvorbis routines and structures are declared in "vorbis/codec.h".</p>
<h2>Encoding workflow</h2>
<ol>
<li>Initialize a <a href="vorbis_info.html">vorbis_info</a> structure
by calling <a href="vorbis_info_init.html">vorbis_info_init</a> and
then functions from <a href="../vorbisenc/index.html">libvorbisenc</a>
on it.</li>
<li>Initialize a <a href="vorbis_dsp_state.html">vorbis_dsp_state</a>
for encoding based on the parameters in the vorbis_info by using <a
href="vorbis_analysis_init.html">vorbis_analysis_init</a>.</li>
<li>Initialize a <a href="vorbis_comment.html">vorbis_comment</a>
structure using <a href="vorbis_comment_init.html">vorbis_comment_init</a>,
populate it with any comments you wish to store in the stream, and call
<a href="vorbis_analysis_headerout.html">vorbis_analysis_headerout</a> to
get the three Vorbis stream header packets. Output the packets.</li>
<li>Initialize a <a href="vorbis_block.html">vorbis_block</a> structure
using <a href="vorbis_block_init.html">vorbis_block_init</a>.</li>
<li>While there is more audio to encode:<ol>
<li>Submit a chunk of audio data using <a
href="vorbis_analysis_buffer.html">vorbis_analysis_buffer</a> and <a
href="vorbis_analysis_wrote.html">vorbis_analysis_wrote</a>.</li>
<li>Obtain all available blocks using <a
href="vorbis_analysis_blockout.html">vorbis_analysis_blockout</a>
in a loop. For each block obtained:<ol>
<li>Encode the block into a packet (or prepare it for bitrate management)
using <a href="vorbis_analysis.html">vorbis_analysis</a>. (It's a good
idea to always pass the blocks through the bitrate
management mechanism; more information is on the <a
href="vorbis_analysis.html">vorbis_analysis</a> page. It does not affect
the resulting packets unless you are actually using a bitrate-managed
mode.)</li>
<li>If you are using bitrate management, submit the block using <a
href="vorbis_bitrate_addblock.html">vorbis_bitrate_addblock</a> and obtain
packets using <a
href="vorbis_bitrate_flushpacket.html">vorbis_bitrate_flushpacket</a>.</li>
<li>Output any obtained packets.</li>
</ol></li>
</ol></li>
<li>Submit an empty buffer to indicate the end of input; this will result
in an end-of-stream packet after all encoding steps are done to it.</li>
<li>Destroy the structures using the appropriate vorbis_*_clear routines.</li>
</ol>
<h2>Decoding workflow</h2>
<em>Note: if you do not need to do anything more involved than just
decoding the audio from an Ogg Vorbis file, you can use the far simpler
<a href="../vorbisfile/index.html">libvorbisfile</a> interface, which
will take care of all of the demuxing and low-level decoding operations
(and even the I/O, if you want) for you.</em>
<ol>
<li>When reading the header packets of an Ogg stream, you can use <a
href="vorbis_synthesis_idheader.html">vorbis_synthesis_idheader</a> to
check whether a stream might be Vorbis.</li>
<li>Initialize a <a href="vorbis_info.html">vorbis_info</a> and a <a
href="vorbis_comment.html">vorbis_comment</a> structure using the
appropriate vorbis_*_init routines, then pass the first three packets
from the stream (the Vorbis stream header packets) to <a
href="vorbis_synthesis_headerin.html">vorbis_synthesis_headerin</a> in
order. At this point, you can see the comments and basic parameters of
the Vorbis stream.</li>
<li>Initialize a <a href="vorbis_dsp_state.html">vorbis_dsp_state</a>
for decoding based on the parameters in the vorbis_info by using <a
href="vorbis_synthesis_init.html">vorbis_synthesis_init</a>.</li>
<li>Initialize a <a href="vorbis_block.html">vorbis_block</a> structure
using <a href="vorbis_block_init.html">vorbis_block_init</a>.</li>
<li>While there are more packets to decode:<ol>
<li>Decode the next packet into a block using <a
href="vorbis_synthesis.html">vorbis_synthesis</a>.</li>
<li>Submit the block to the reassembly layer using <a
href="vorbis_synthesis_blockin.html">vorbis_synthesis_blockin</a>.</li>
<li>Obtain some decoded audio using <a
href="vorbis_synthesis_pcmout.html">vorbis_synthesis_pcmout</a> and <a
href="vorbis_synthesis_read.html">vorbis_synthesis_read</a>. Any audio data
returned but not marked as consumed using vorbis_synthesis_read carries
over to the next call to vorbis_synthesis_pcmout.</li>
</ol></li>
<li>Destroy the structures using the appropriate vorbis_*_clear routines.</li>
</ol>
<h2>Metadata workflow</h2>
<em>Note: if you do not need to do anything more involved than just
reading the metadata from an Ogg Vorbis file, <a
href="../vorbisfile/index.html">libvorbisfile</a> can do this for you.</em>
<ol>
<li>Follow the decoding workflow above until you have access to the comments
and basic parameters of the Vorbis stream.</li>
<li>If you want to alter the comments, copy the first packet to the output
file, then create a packet for the modified comments using <a
href="vorbis_commentheader_out.html">vorbis_commentheader_out</a> and output
it, then copy the third packet and all subsequent packets into the output
file.</li>
</ol>
<br><br>
<hr noshade>
<table border=0 width=100%>
<tr valign=top>
<td><p class=tiny>copyright &copy; 2010 Xiph.Org</p></td>
<td align=right><p class=tiny><a href="http://www.xiph.org/ogg/vorbis/index.html">Ogg Vorbis</a></p></td>
</tr><tr>
<td><p class=tiny>libvorbis documentation</p></td>
<td align=right><p class=tiny>libvorbis version 1.3.2 - 20101101</p></td>
</tr>
</table>
</body>
</html>