ShuriZma
/
pcsx2
mirror of https://github.com/PCSX2/pcsx2.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
pcsx2/3rdparty/SoundTouch/README.html

753 lines
30 KiB
HTML
Raw Normal View History

Updated the SoundTouch library to version 1.5. It works fine and without further modifications in Windows, for SPU2-X and ZeroSPU2. Linux is untested, but hopefully also fine :p git-svn-id: http://pcsx2.googlecode.com/svn/trunk@2734 96395faa-99c1-11dd-bbfe-3dabce05a288
2010-03-18 20:04:51 +00:00
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type"
content="text/html; charset=windows-1252">
<meta http-equiv="Content-Language" content="en-us">
<meta name="author" content="Olli Parviainen">
<meta name="description"
content="Readme file for SoundTouch audio processing library">
<meta name="GENERATOR" content="Microsoft FrontPage 4.0">
<meta name="ProgId" content="FrontPage.Editor.Document">
<title>SoundTouch library README</title>
<style>
<!--
.normal { font-family: Arial }
-->
</style>
</head>
<body class="normal">
<hr>
<h1>SoundTouch audio processing library v1.5.0
</h1>
<p class="normal">SoundTouch library Copyright (c) Olli
Parviainen 2002-2009 </p>
<hr>
<h2>1. Introduction </h2>
<p>SoundTouch is an open-source audio
processing library that allows changing the sound tempo, pitch
and playback rate parameters independently from each other, i.e.:</p>
<ul>
<li>Sound tempo can be increased or decreased while
maintaining the original pitch</li>
<li>Sound pitch can be increased or decreased while
maintaining the original tempo </li>
<li>Change playback rate that affects both tempo
and pitch at the same time </li>
<li>Choose any combination of tempo/pitch/rate</li>
</ul>
<h3>1.1 Contact information </h3>
<p>Author email: oparviai 'at' iki.fi </p>
<p>SoundTouch WWW page: <a href="http://www.surina.net/soundtouch">http://www.surina.net/soundtouch</a></p>
<hr>
<h2>2. Compiling SoundTouch</h2>
<p>Before compiling, notice that you can choose the sample data format
if it's desirable to use floating point sample
data instead of 16bit integers. See section "sample data format"
for more information.</p>
<h3>2.1. Building in Microsoft Windows</h3>
<p>Project files for Microsoft Visual C++ 6.0 and Visual C++ .NET are
supplied with the source code package.&nbsp;</p>
<p> Please notice that SoundTouch
library uses processor-specific optimizations for Pentium III and AMD
processors. Visual Studio .NET and later versions supports the required
instructions by default, but Visual Studio 6.0 requires a processor pack upgrade
to be installed in order to support these optimizations. The processor pack upgrade can be downloaded from
Microsoft site at this URL:</p>
<p><a href="http://msdn.microsoft.com/en-us/vstudio/aa718349.aspx">http://msdn.microsoft.com/en-us/vstudio/aa718349.aspx</a></p>
<p>If the above URL is unavailable or removed, go
to <a href="http://msdn.microsoft.com/">http://msdn.microsoft.com</a>
and perform a search with keywords &quot;processor pack&quot;. </p>
<p>To build the binaries with Visual C++
compiler, either run &quot;make-win.bat&quot; script, or open the
appropriate project files in source code directories with Visual
Studio. The final executable will appear under the &quot;SoundTouch\bin&quot;
directory. If using the Visual Studio IDE instead of the make-win.bat script, directories bin and
lib may need to be created manually to the SoundTouch
package root for the final executables. The make-win.bat script
creates these directories automatically.
</p>
<h3>2.2. Building in Gnu platforms</h3>
<p>The SoundTouch library can be compiled in
practically any platform supporting GNU compiler (GCC) tools.
SoundTouch have been tested with gcc version 3.3.4., but it
shouldn't be very specific about the gcc version. Assembler-level
performance optimizations for GNU platform are currently available in
x86 platforms only, they are automatically disabled and replaced with
standard C routines in other processor platforms.</p>
<p>To build and install the binaries, run the
following commands in the SoundTouch/ directory:</p>
<table border="0" cellpadding="0" cellspacing="4">
<tbody>
<tr valign="top">
<td>
<pre>./configure -</pre>
</td>
<td>
<p>Configures the SoundTouch package for the local
environment.</p>
</td>
</tr>
<tr valign="top">
<td>
<pre>make -</pre>
</td>
<td>
<p>Builds the SoundTouch library &amp;
SoundStretch utility.</p>
</td>
</tr>
<tr valign="top">
<td>
<pre>make install -</pre>
</td>
<td>
<p>Installs the SoundTouch &amp; BPM libraries
to <b>/usr/local/lib</b> and SoundStretch utility to <b>/usr/local/bin</b>.
Please notice that 'root' privileges may be required to install the
binaries to the destination locations.</p>
</td>
</tr>
</tbody>
</table>
<h4><b>2.2.1 Required GNU tools</b>&nbsp;</h4>
<p> Bash shell, GNU C++ compiler, libtool, autoconf and automake tools are required
for compiling
the SoundTouch library. These are usually included with the GNU/Linux distribution, but if
not, install these packages first. For example, in Ubuntu Linux these can be acquired and
installed with the following command:</p>
<pre><b>sudo apt-get install <font SIZE="2">automake autoconf libtool build-essential</font></b></pre>
<h4><b>2.2.2 Problems with GCC compiler compatibility</b></h4>
<p>At the release time the SoundTouch package has been tested to compile in
GNU/Linux platform. However, in past it's happened that new gcc versions aren't
necessarily compatible with the assembler settings used in the optimized
routines. <b>If you have problems getting the
SoundTouch library compiled, try the workaround of disabling the optimizations</b>
by editing the file &quot;include/STTypes.h&quot; and removing the following
definition there:</p>
<blockquote>
<pre>#define ALLOW_OPTIMIZATIONS 1</pre>
</blockquote>
<h4><b>2.2.3 Problems with configure script or build process</b>&nbsp;</h4>
<p>Incompatibilities between various GNU toolchain versions may cause errors when running the &quot;configure&quot; script or building the source
codes, if your GNU tool versions are not compatible with the versions used for
preparing the SoundTouch kit.&nbsp;</p>
<p>To resolve the issue, regenerate the configure scripts with your local tool
set by running
the &quot;<b>./bootstrap</b>&quot; script included in the SoundTouch source code
kit. After that, run the <b>configure</b> script and <b>make</b> as usually.</p>
<h4><b>2.2.4 Compiler issues with non-x86 processors</b></h4>
<p>SoundTouch library works also on non-x86 processors.</p>
<p>However, in case that you get compiler errors when trying to compile for non-Intel processor, edit the file
&quot;<b>source\SoundTouch\Makefile.am</b>&quot; and remove the &quot;<b>-msse2</b>&quot;
flag on the <b>AM_CXXFLAGS </b>line:</p>
<pre><b>AM_CXXFLAGS=-O3 -fcheck-new -I../../include&nbsp;&nbsp;&nbsp; # Note: -msse2 flag removed!</b></pre>
<p>After that, run &quot;<b>./bootstrap</b>&quot; script, and then run <b>configure</b>
and <b>make</b> again.</p>
<hr>
<h2>3. About implementation &amp; Usage tips</h2>
<h3>3.1. Supported sample data formats</h3>
<p>The sample data format can be chosen
between 16bit signed integer and 32bit floating point values, the
default is 32bit floating point. </p>
<p>
In Windows environment, the sample data format is chosen
in file &quot;STTypes.h&quot; by choosing one of the following
defines:</p>
<ul>
<li><span style="font-weight: bold;">#define INTEGER_SAMPLES</span>
for 16bit signed
integer</li>
<li><span style="font-weight: bold;">#define FLOAT_SAMPLES</span> for
32bit floating point</li>
</ul>
<p>
In GNU environment, the floating sample format is used by default, but
integer sample format can be chosen by giving the
following switch to the configure script:
<blockquote>
<pre>./configure --enable-integer-samples</pre>
</blockquote>
<p>The sample data can have either single (mono)
or double (stereo) audio channel. Stereo data is interleaved so
that every other data value is for left channel and every second
for right channel. Notice that while it'd be possible in theory
to process stereo sound as two separate mono channels, this isn't
recommended because processing the channels separately would
result in losing the phase coherency between the channels, which
consequently would ruin the stereo effect.</p>
<p>Sample rates between 8000-48000H are
supported.</p>
<h3>3.2. Processing latency</h3>
<p>The processing and latency constraints of
the SoundTouch library are:</p>
<ul>
<li>Input/output processing latency for the
SoundTouch processor is around 100 ms. This is when time-stretching is
used. If the rate transposing effect alone is used, the latency
requirement
is much shorter, see section 'About algorithms'.</li>
<li>Processing CD-quality sound (16bit stereo
sound with 44100H sample rate) in real-time or faster is possible
starting from processors equivalent to Intel Pentium 133Mh or better,
if using the "quick" processing algorithm. If not using the "quick"
mode or
if floating point sample data are being used, several times more CPU
power is typically required.</li>
</ul>
<h3>3.3. About algorithms</h3>
<p>SoundTouch provides three seemingly
independent effects: tempo, pitch and playback rate control.
These three controls are implemented as combination of two primary
effects, <em>sample rate transposing</em> and <em>time-stretching</em>.</p>
<p><em>Sample rate transposing</em> affects
both the audio stream duration and pitch. It's implemented simply
by converting the original audio sample stream to the&nbsp; desired
duration by interpolating from the original audio samples. In SoundTouch, linear interpolation with anti-alias filtering is
used. Theoretically a higher-order interpolation provide better
result than 1st order linear interpolation, but in audio
application linear interpolation together with anti-alias
filtering performs subjectively about as well as higher-order
filtering would.</p>
<p><em>Time-stretching </em>means changing
the audio stream duration without affecting it's pitch. SoundTouch
uses WSOLA-like time-stretching routines that operate in the time
domain. Compared to sample rate transposing, time-stretching is a
much heavier operation and also requires a longer processing
"window" of sound samples used by the
processing algorithm, thus increasing the algorithm input/output
latency. Typical i/o latency for the SoundTouch
time-stretch algorithm is around 100 ms.</p>
<p>Sample rate transposing and time-stretching
are then used together to produce the tempo, pitch and rate
controls:</p>
<ul>
<li><strong>'Tempo'</strong> control is
implemented purely by time-stretching.</li>
<li><strong>'Rate</strong>' control is implemented
purely by sample rate transposing.</li>
<li><strong>'Pitch</strong>' control is
implemented as a combination of time-stretching and sample rate
transposing. For example, to increase pitch the audio stream is first
time-stretched to longer duration (without affecting pitch) and then
transposed back to original duration by sample rate transposing, which
simultaneously reduces duration and increases pitch. The result is
original duration but increased pitch.</li>
</ul>
<h3>3.4 Tuning the algorithm parameters</h3>
<p>The time-stretch algorithm has few
parameters that can be tuned to optimize sound quality for
certain application. The current default parameters have been
chosen by iterative if-then analysis (read: "trial and error")
to obtain best subjective sound quality in pop/rock music
processing, but in applications processing different kind of
sound the default parameter set may result into a sub-optimal
result.</p>
<p>The time-stretch algorithm default
parameter values are set by the following #defines in file &quot;TDStretch.h&quot;:</p>
<blockquote>
<pre>#define DEFAULT_SEQUENCE_MS AUTOMATIC
#define DEFAULT_SEEKWINDOW_MS AUTOMATIC
#define DEFAULT_OVERLAP_MS 8</pre>
</blockquote>
<p>These parameters affect to the time-stretch
algorithm as follows:</p>
<ul>
<li><strong>DEFAULT_SEQUENCE_MS</strong>: This is
the default length of a single processing sequence in milliseconds
which determines the how the original sound is chopped in
the time-stretch algorithm. Larger values mean fewer sequences
are used in processing. In principle a larger value sounds better when
slowing down the tempo, but worse when increasing the tempo and vice
versa.&nbsp;<br>
<br>
By default, this setting value is calculated automatically according to
tempo value.<br>
</li>
<li><strong>DEFAULT_SEEKWINDOW_MS</strong>: The seeking window
default length in milliseconds is for the algorithm that seeks the best
possible overlapping location. This determines from how
wide a sample "window" the algorithm can use to find an optimal mixing
location when the sound sequences are to be linked back together.&nbsp;<br>
<br>
The bigger this window setting is, the higher the possibility to find a
better mixing position becomes, but at the same time large values may
cause a "drifting" sound artifact because neighboring sequences can be
chosen at more uneven intervals. If there's a disturbing artifact that
sounds as if a constant frequency was drifting around, try reducing
this setting.<br>
<br>
By default, this setting value is calculated automatically according to
tempo value.<br>
</li>
<li><strong>DEFAULT_OVERLAP_MS</strong>: Overlap
length in milliseconds. When the sound sequences are mixed back
together to form again a continuous sound stream, this parameter
defines how much the ends of the consecutive sequences will overlap with each other.<br>
<br>
This shouldn't be that critical parameter. If you reduce the
DEFAULT_SEQUENCE_MS setting by a large amount, you might wish to try a
smaller value on this.</li>
</ul>
<p>Notice that these parameters can also be
set during execution time with functions "<strong>TDStretch::setParameters()</strong>"
and "<strong>SoundTouch::setSetting()</strong>".</p>
<p>The table below summaries how the
parameters can be adjusted for different applications:</p>
<table border="1">
<tbody>
<tr>
<td valign="top"><strong>Parameter name</strong></td>
<td valign="top"><strong>Default value
magnitude</strong></td>
<td valign="top"><strong>Larger value
affects...</strong></td>
<td valign="top"><strong>Smaller value
affects...</strong></td>
<td valign="top"><strong>Effect to CPU burden</strong></td>
</tr>
<tr>
<td valign="top">
<pre>SEQUENCE_MS</pre>
</td>
<td valign="top">Default value is relatively
large, chosen for slowing down music tempo</td>
<td valign="top">Larger value is usually
better for slowing down tempo. Growing the value decelerates the
"echoing" artifact when slowing down the tempo.</td>
<td valign="top">Smaller value might be better
for speeding up tempo. Reducing the value accelerates the "echoing"
artifact when slowing down the tempo </td>
<td valign="top">Increasing the parameter
value reduces computation burden</td>
</tr>
<tr>
<td valign="top">
<pre>SEEKWINDOW_MS</pre>
</td>
<td valign="top">Default value is relatively
large, chosen for slowing down music tempo</td>
<td valign="top">Larger value eases finding a
good mixing position, but may cause a "drifting" artifact</td>
<td valign="top">Smaller reduce possibility to
find a good mixing position, but reduce the "drifting" artifact.</td>
<td valign="top">Increasing the parameter
value increases computation burden</td>
</tr>
<tr>
<td valign="top">
<pre>OVERLAP_MS</pre>
</td>
<td valign="top">Default value is relatively
large, chosen to suit with above parameters.</td>
<td valign="top">&nbsp;</td>
<td valign="top">If you reduce the "sequence
ms" setting, you might wish to try a smaller value.</td>
<td valign="top">Increasing the parameter
value increases computation burden</td>
</tr>
</tbody>
</table>
<h3>3.5 Performance Optimizations </h3>
<p><strong>General optimizations:</strong></p>
<p>The time-stretch routine has a 'quick' mode
that substantially speeds up the algorithm but may degrade the
sound quality by a small amount. This mode is activated by
calling SoundTouch::setSetting() function with parameter&nbsp; id
of SETTING_USE_QUICKSEEK and value "1", i.e. </p>
<blockquote>
<p>setSetting(SETTING_USE_QUICKSEEK, 1);</p>
</blockquote>
<p><strong>CPU-specific optimizations:</strong></p>
<ul>
<li>Intel MMX optimized routines are used with
compatible CPUs when 16bit integer sample type is used. MMX optimizations are available both in Win32 and Gnu/x86 platforms.
Compatible processors are Intel PentiumMMX and later; AMD K6-2, Athlon
and later. </li>
<li>Intel SSE optimized routines are used with
compatible CPUs when floating point sample type is used. SSE optimizations are currently implemented for Win32 platform only.
Processors compatible with SSE extension are Intel processors starting
from Pentium-III, and AMD processors starting from Athlon XP. </li>
<li>AMD 3DNow! optimized routines are used with
compatible CPUs when floating point sample type is used, but SSE
extension isn't supported . 3DNow! optimizations are currently
implemented for Win32 platform only. These optimizations are used in
AMD K6-2 and Athlon (classic) CPU's; better performing SSE routines are
used with AMD processor starting from Athlon XP. </li>
</ul>
<hr>
<h2><a name="SoundStretch"></a>4. SoundStretch audio processing utility
</h2>
<p>SoundStretch audio processing utility<br>
Copyright (c) Olli Parviainen 2002-2009</p>
<p>SoundStretch is a simple command-line
application that can change tempo, pitch and playback rates of
WAV sound files. This program is intended primarily to
demonstrate how the &quot;SoundTouch&quot; library can be used to
process sound in your own program, but it can as well be used for
processing sound files.</p>
<h3>4.1. SoundStretch Usage Instructions</h3>
<p>SoundStretch Usage syntax:</p>
<blockquote>
<pre>soundstretch infilename outfilename [switches]</pre>
</blockquote>
<p>Where: </p>
<table border="0" cellpadding="2" width="100%">
<tbody>
<tr>
<td valign="top">
<pre>&quot;infilename&quot;</pre>
</td>
<td valign="top">Name of the input sound
data file (in .WAV audio file format). Give &quot;stdin&quot; as filename to use
standard input pipe. </td>
</tr>
<tr>
<td valign="top">
<pre>&quot;outfilename&quot;</pre>
</td>
<td valign="top">Name of the output sound
file where the resulting sound is saved (in .WAV audio file format).
This parameter may be omitted if you&nbsp; don't want to save the
output
(e.g. when only calculating BPM rate with '-bpm' switch). Give &quot;stdout&quot;
as filename to use standard output pipe.</td>
</tr>
<tr>
<td valign="top">
<pre>&nbsp;[switches]</pre>
</td>
<td valign="top">Are one or more control
switches.</td>
</tr>
</tbody>
</table>
<p>Available control switches are:</p>
<table border="0" cellpadding="2" width="100%">
<tbody>
<tr>
<td valign="top">
<pre>-tempo=n </pre>
</td>
<td valign="top">Change the sound tempo by n
percents (n = -95.0 .. +5000.0 %) </td>
</tr>
<tr>
<td valign="top">
<pre>-pitch=n</pre>
</td>
<td valign="top">Change the sound pitch by n
semitones (n = -60.0 .. + 60.0 semitones) </td>
</tr>
<tr>
<td valign="top">
<pre>-rate=n</pre>
</td>
<td valign="top">Change the sound playback rate by
n percents (n = -95.0 .. +5000.0 %) </td>
</tr>
<tr>
<td valign="top">
<pre>-bpm=n</pre>
</td>
<td valign="top">Detect the Beats-Per-Minute (BPM) rate of the sound and adjust the tempo to meet 'n'
BPMs. When this switch is
applied, the &quot;-tempo&quot; switch is ignored. If "=n" is
omitted, i.e. switch &quot;-bpm&quot; is used alone, then the BPM rate is
estimated and displayed, but tempo not adjusted according to the BPM
value. </td>
</tr>
<tr>
<td valign="top">
<pre>-quick</pre>
</td>
<td valign="top">Use quicker tempo change
algorithm. Gains speed but loses sound quality. </td>
</tr>
<tr>
<td valign="top">
<pre>-naa</pre>
</td>
<td valign="top">Don't use anti-alias
filtering in sample rate transposing. Gains speed but loses sound
quality. </td>
</tr>
<tr>
<td valign="top">
<pre>-license</pre>
</td>
<td valign="top">Displays the program license
text (LGPL)</td>
</tr>
</tbody>
</table>
<p>Notes:</p>
<ul>
<li>To use standard input/output pipes for processing, give &quot;stdin&quot;
and &quot;stdout&quot; as input/output filenames correspondingly. The
standard input/output pipes will still carry the audio data in .wav audio
file format.</li>
<li>The numerical switches allow both integer (e.g. "-tempo=123") and decimal (e.g.
"-tempo=123.45") numbers.</li>
<li>The &quot;-naa&quot; and/or "-quick" switches can be
used to reduce CPU usage while compromising some sound quality </li>
<li>The BPM detection algorithm works by detecting
repeating bass or drum patterns at low frequencies of &lt;250Hz. A
lower-than-expected BPM figure may be reported for music with uneven or
complex bass patterns. </li>
</ul>
<h3>4.2. SoundStretch usage examples </h3>
<p><strong>Example 1</strong></p>
<p>The following command increases tempo of
the sound file &quot;originalfile.wav&quot; by 12.5% and stores result to file &quot;destinationfile.wav&quot;:</p>
<blockquote>
<pre>soundstretch originalfile.wav destinationfile.wav -tempo=12.5</pre>
</blockquote>
<p><strong>Example 2</strong></p>
<p>The following command decreases the sound
pitch (key) of the sound file &quot;orig.wav&quot; by two
semitones and stores the result to file &quot;dest.wav&quot;:</p>
<blockquote>
<pre>soundstretch orig.wav dest.wav -pitch=-2</pre>
</blockquote>
<p><strong>Example 3</strong></p>
<p>The following command processes the file &quot;orig.wav&quot; by decreasing the sound tempo by 25.3% and
increasing the sound pitch (key) by 1.5 semitones. Resulting .wav audio data is
directed to standard output pipe:</p>
<blockquote>
<pre>soundstretch orig.wav stdout -tempo=-25.3 -pitch=1.5</pre>
</blockquote>
<p><strong>Example 4</strong></p>
<p>The following command detects the BPM rate
of the file &quot;orig.wav&quot; and adjusts the tempo to match
100 beats per minute. Result is stored to file &quot;dest.wav&quot;:</p>
<blockquote>
<pre>soundstretch orig.wav dest.wav -bpm=100</pre>
</blockquote>
<p><strong>Example 5</strong></p>
<p>The following command reads .wav sound data from standard input pipe and
estimates the BPM rate:</p>
<blockquote>
<pre>soundstretch stdin -bpm</pre>
</blockquote>
<hr>
<h2>5. Change History</h2>
<h3>5.1. SoundTouch library Change History </h3>
<p><strong>1.5.0:</strong></p>
<ul>
<li>Added normalization to correlation calculation and improvement automatic seek/sequence parameter calculation to improve sound quality</li>
<li>Bugfixes:&nbsp;
<ul>
<li>Fixed negative array indexing in quick seek algorithm</li>
<li>FIR autoalias filter running too far in processing buffer</li>
<li>Check against zero sample count in rate transposing</li>
<li>Fix for x86-64 support: Removed pop/push instructions from the cpu detection algorithm.&nbsp;</li>
<li>Check against empty buffers in FIFOSampleBuffer</li>
<li>Other minor fixes &amp; code cleanup</li>
</ul>
</li>
<li>Fixes in compilation scripts for non-Intel platforms</li>
<li>Added Dynamic-Link-Library (DLL) version of SoundTouch library build,
provided with Delphi/Pascal wrapper for calling the dll routines</li>
<li>Added #define PREVENT_CLICK_AT_RATE_CROSSOVER that prevents a click artifact
when crossing the nominal pitch from either positive to negative side or vice
versa</li>
</ul>
<p><strong>1.4.1:</strong></p>
<ul>
<li>Fixed a buffer overflow bug in BPM detect algorithm routines if processing
more than 2048 samples at one call&nbsp;</li>
</ul>
<p><strong>1.4.0:</strong></p>
<ul>
<li>Improved sound quality by automatic calculation of time stretch algorithm
processing parameters according to tempo setting</li>
<li>Moved BPM detection routines from SoundStretch application into SoundTouch
library</li>
<li>Bugfixes: Usage of uninitialied variables, GNU build scripts, compiler errors
due to 'const' keyword mismatch.</li>
<li>Source code cleanup</li>
</ul>
<p><strong>v1.3.1:
</strong></p>
<ul>
<li>Changed static class declaration to GCC 4.x compiler compatible syntax.</li>
<li>Enabled MMX/SSE-optimized routines also for GCC compilers. Earlier
the MMX/SSE-optimized routines were written in compiler-specific inline
assembler, now these routines are migrated to use compiler intrinsic
syntax which allows compiling the same MMX/SSE-optimized source code with
both Visual C++ and GCC compilers. </li>
<li>Set floating point as the default sample format and added switch to
the GNU configure script for selecting the other sample format.</li>
</ul>
<p><strong>v1.3.0:
</strong></p>
<ul>
<li>Fixed tempo routine output duration inaccuracy due to rounding
error </li>
<li>Implemented separate processing routines for integer and
floating arithmetic to allow improvements to floating point routines
(earlier used algorithms mostly optimized for integer arithmetic also
for floating point samples) </li>
<li>Fixed a bug that distorts sound if sample rate changes during the
sound stream </li>
<li>Fixed a memory leak that appeared in MMX/SSE/3DNow! optimized
routines </li>
<li>Reduced redundant code pieces in MMX/SSE/3DNow! optimized
routines vs. the standard C routines.</li>
<li>MMX routine incompatibility with new gcc compiler versions </li>
<li>Other miscellaneous bug fixes </li>
</ul>
<p><strong>v1.2.1: </strong></p>
<ul>
<li>Added automake/autoconf scripts for GNU
platforms (in courtesy of David Durham)</li>
<li>Fixed SCALE overflow bug in rate transposer
routine.</li>
<li>Fixed 64bit address space bugs.</li>
<li>Created a 'soundtouch' namespace for
SAMPLETYPE definitions.</li>
</ul>
<p><strong>v1.2.0: </strong></p>
<ul>
<li>Added support for 32bit floating point sample
data type with SSE/3DNow! optimizations for Win32 platform (SSE/3DNow! optimizations currently not supported in GCC environment)</li>
<li>Replaced 'make-gcc' script for GNU environment
by master Makefile</li>
<li>Added time-stretch routine configurability to
SoundTouch main class</li>
<li>Bugfixes</li>
</ul>
<p><strong>v1.1.1: </strong></p>
<ul>
<li>Moved SoundTouch under lesser GPL license (LGPL). This allows using SoundTouch library in programs that aren't
released under GPL license. </li>
<li>Changed MMX routine organiation so that MMX optimized routines are now implemented in classes that are derived from
the basic classes having the standard non-mmx routines. </li>
<li>MMX routines to support gcc version 3. </li>
<li>Replaced windows makefiles by script using the .dsw files </li>
</ul>
<p><strong>v1.01: </strong></p>
<ul>
<li>&quot;mmx_gcc.cpp&quot;: Added "using namespace std" and
removed "return 0" from a function with void return value to fix
compiler errors when compiling the library in Solaris environment. </li>
<li>Moved file &quot;FIFOSampleBuffer.h&quot; to "include"
directory to allow accessing the FIFOSampleBuffer class from external
files. </li>
</ul>
<p><strong>v1.0: </strong></p>
<ul>
<li>Initial release </li>
</ul>
<p>&nbsp;</p>
<h3>5.2. SoundStretch application Change
History </h3>
<p><strong>1.4.0:</strong></p>
<ul>
<li>Moved BPM detection routines from SoundStretch application into SoundTouch
library</li>
<li>Allow using standard input/output pipes as audio processing input/output
streams</li>
</ul>
<p><strong>v1.3.0:</strong></p>
<ul>
<li>Simplified accessing WAV files with floating
point sample format.
</li>
</ul>
<p><strong>v1.2.1: </strong></p>
<ul>
<li>Fixed 64bit address space bugs.</li>
</ul>
<p><strong>v1.2.0: </strong></p>
<ul>
<li>Added support for 32bit floating point sample
data type</li>
<li>Restructured the BPM routines into separate
library</li>
<li>Fixed big-endian conversion bugs in WAV file
routines (hopefully :)</li>
</ul>
<p><strong>v1.1.1: </strong></p>
<ul>
<li>Fixed bugs in WAV file reading &amp; added
byte-order conversion for big-endian processors. </li>
<li>Moved SoundStretch source code under 'example'
directory to highlight difference from SoundTouch stuff. </li>
<li>Replaced windows makefiles by script using the .dsw files </li>
<li>Output file name isn't required if output
isn't desired (e.g. if using the switch '-bpm' in plain format only) </li>
</ul>
<p><strong>v1.1:</strong></p>
<ul>
<li>Fixed "Release" settings in Microsoft Visual
C++ project file (.dsp) </li>
<li>Added beats-per-minute (BPM) detection routine
and command-line switch &quot;-bpm&quot; </li>
</ul>
<p><strong>v1.01: </strong></p>
<ul>
<li>Initial release </li>
</ul>
<hr>
<h2 >6. Acknowledgements </h2>
<p >Kudos for these people who have contributed to development or submitted
bugfixes since
SoundTouch v1.3.1: </p>
<ul>
<li>Arthur A</li>
<li>Richard Ash</li>
<li>Stanislav Brabec</li>
<li>Christian Budde</li>
<li>Brian Cameron</li>
<li>Jason Champion</li>
<li>Patrick Colis</li>
<li>Justin Frankel</li>
<li>Jason Garland</li>
<li>Takashi Iwai</li>
<li>Paulo Pizarro</li>
<li>RJ Ryan</li>
<li>John Sheehy</li>
</ul>
<p >Moral greetings to all other contributors and users also!</p>
<hr>
<h2 >7. LICENSE </h2>
<p>SoundTouch audio processing library<br>
Copyright (c) Olli Parviainen</p>
<p>This library is free software; you can
redistribute it and/or modify it under the terms of the GNU
Lesser General Public License version 2.1 as published by the Free Software
Foundation.</p>
<p>This library is distributed in the hope
that it will be useful, but WITHOUT ANY WARRANTY; without even
the implied warranty of MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE. See the GNU Lesser General Public License for
more details.</p>
<p>You should have received a copy of the GNU
Lesser General Public License along with this library; if not,
write to the Free Software Foundation, Inc., 59 Temple Place,
Suite 330, Boston, MA 02111-1307 USA</p>
<hr>
<!--
$Id: README.html 81 2009-12-28 20:51:18Z oparviai $
-->
</body>
</html>