diff options
Diffstat (limited to 'lib/soundtouch/README.html')
-rw-r--r-- | lib/soundtouch/README.html | 923 |
1 files changed, 0 insertions, 923 deletions
diff --git a/lib/soundtouch/README.html b/lib/soundtouch/README.html deleted file mode 100644 index 84a5345831..0000000000 --- a/lib/soundtouch/README.html +++ /dev/null @@ -1,923 +0,0 @@ -<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> -<html> -<head> - <title>SoundTouch library README</title> - <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"> - <style> - body {font-family: Arial, Helvetica; } - </style> -</head> -<body class="normal"> -<hr> -<h1>SoundTouch audio processing library v2.1.1</h1> -<p class="normal">SoundTouch library Copyright © Olli Parviainen 2001-2018</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://soundtouch.surina.net">http://soundtouch.surina.net</a></p> -<p>SoundTouch git repository: <a href="https://gitlab.com/soundtouch/soundtouch.git">https://gitlab.com/soundtouch/soundtouch.git</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> -<p>Also notice that SoundTouch can use OpenMP instructions for parallel -computation to accelerate the runtime processing speed in multi-core systems, -however, these improvements need to be separately enabled before compiling. See -OpenMP notes in Chapter 3 below.</p> -<h3>2.1. Building in Microsoft Windows</h3> -<p>Project files for Microsoft Visual C++ are supplied with the source -code package. Go to Microsoft WWW page to download -<a href="http://www.visualstudio.com/en-US/products/visual-studio-express-vs"> -Microsoft Visual Studio Express version for free</a>. -</p> -<p>To build the binaries with Visual C++ compiler, either run -"make-win.bat" script, or open the appropriate project files in source -code directories with Visual Studio. The final executable will appear -under the "SoundTouch\bin" 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> -<p><strong>C# example</strong>: The source code package includes also a C# example - application for Windows that shows how to invoke SoundTouch.dll - dynamic-load library for processing mp3 audio. -<p><strong>OpenMP NOTE</strong>: If activating the OpenMP parallel computing in -the compilation, the target program will require additional vcomp dll library to -properly run. In Visual C++ 9.0 these libraries can be found in the following -folders.</p> -<ul> - <li>x86 32bit: C:\Program Files (x86)\Microsoft Visual Studio - 9.0\VC\redist\x86\Microsoft.VC90.OPENMP\vcomp90.dll</li> - <li>x64 64bit: C:\Program Files (x86)\Microsoft Visual Studio - 9.0\VC\redist\amd64\Microsoft.VC90.OPENMP\vcomp90.dll</li> -</ul> -<p>In Visual Studio 2008, a SP1 version may be required for these libraries. In -other VC++ versions the required library will be expectedly found in similar -"redist" location.</p> -<p>Notice that as minor demonstration of a "dll hell" phenomenon both the 32-bit -and 64-bit version of vcomp90.dll have the same filename but different contents, -thus choose the proper version to allow the program start.</p> -<h3>2.2. Building in Gnu platforms</h3> -<p>The SoundTouch library compiles in practically any platform -supporting GNU compiler (GCC) tools. SoundTouch requires GCC version 4.3 or later.</p> -<p>To build and install the binaries, run the following commands in -/soundtouch directory:</p> -<table border="0" cellpadding="0" cellspacing="4"> - <tbody> - <tr> - <td style="vertical-align: top;"> - <pre>./bootstrap -</pre> - </td> - <td style="vertical-align: top;">Creates "configure" file with -local autoconf/automake toolset.<br> - </td> - </tr> - <tr valign="top"> - <td> - <pre>./configure -</pre> - </td> - <td> - <p>Configures the SoundTouch package for the local environment. -Notice that "configure" file is not available before running the -"./bootstrap" command as above.<br> - </p> - </td> - </tr> - <tr valign="top"> - <td> - <pre>make -</pre> - </td> - <td> - <p>Builds the SoundTouch library & SoundStretch utility. You can - optionally add "-j" switch after "make" to speed up the compilation in - multi-core systems.</p> - </td> - </tr> - <tr valign="top"> - <td> - <pre>make install -</pre> - </td> - <td> - <p>Installs the SoundTouch & 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></h4> -<p> <span style="font-weight: bold;">Bash shell</span>, <span - style="font-weight: bold;">GNU C++ compiler</span>, <span - style="font-weight: bold;">libtool</span>, <span - style="font-weight: bold;">autoconf</span> and <span - style="font-weight: bold;">automake</span> 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, Ubuntu Linux can acquire and install -these with the following command:</p> -<pre><b>sudo apt-get install automake autoconf libtool build-essential</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, If you have problems getting the -SoundTouch library compiled, try disabling optimizations that are specific for -x86 processors by running <b>./configure</b> script with switch -<blockquote> -<pre>--enable-x86-optimizations=no</pre> -</blockquote> - -Alternatively, if you don't use GNU Configure system, edit file "include/STTypes.h" -directly and remove the following definition:<blockquote> - <pre>#define SOUNDTOUCH_ALLOW_X86_OPTIMIZATIONS 1</pre> -</blockquote> - -<h4><b>2.2.3 Compiling Shared Library / DLL version in Cygwin</b></h4> - <p> - The GNU compilation does not automatically create a shared-library version of - SoundTouch (.so or .dll). If such is desired, then you can create it as follows - after running the usual compilation:</p> - <blockquote> - <pre>g++ -shared -static -DDLL_EXPORTS -I../../include -o SoundTouch.dll \ - SoundTouchDLL.cpp ../SoundTouch/.libs/libSoundTouch.a -sstrip SoundTouch.dll</pre> -</blockquote> - -<h3>2.3. Building in Android</h3> -<p>Android compilation instructions are within the - source code package, see file "<b>source/Android-lib/README-SoundTouch-Android.html</b>" - in the source code package. </p> -<p>The Android compilation automatically builds separate .so library binaries -for ARM, X86 and MIPS processor architectures. For optimal device support, -include all these .so library binaries into the Android .apk application -package, so the target Android device can automatically choose the proper -library binary version to use.</p> -<p>The <strong>source/Android-lib</strong> folder includes also an Android -example application that processes WAV audio files using SoundTouch library in -Android devices.</p> - -<hr> -<h2>3. About implementation & Usage tips <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 format, -which will also provide slightly better sound quality over the integer format. </p> -<p> In Windows environment, the sample data format is chosen in file -"STTypes.h" by choosing one of the following defines:</p> -<ul> - <li> <span style="font-weight: bold;">#define -SOUNDTOUCH_INTEGER_SAMPLES</span> for 16bit signed integer</li> - <li> <span style="font-weight: bold;">#define </span><span - style="font-weight: bold;">SOUNDTOUCH_</span><span - style="font-weight: bold;">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: </p> -<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 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 "TDStretch.h":</p> -<blockquote> - <pre>#define DEFAULT_SEQUENCE_MS AUTOMATIC<br>#define DEFAULT_SEEKWINDOW_MS AUTOMATIC<br>#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.<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.<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"></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 slightly compromise the sound quality. -This mode is activated by calling SoundTouch::setSetting() -function with parameter 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> -<p>Intel x86 specific SIMD optimizations are implemented using compiler -intrinsics, providing about a 3x processing speedup for x86 compatible -processors vs. non-SIMD implementation:</p> -<ul> - <li> Intel MMX optimized routines are used with x86 CPUs when 16bit integer - sample type is used</li> - <li> Intel SSE optimized routines are used with x86 CPUs when 32bit floating - point sample type is used</li> -</ul> -<h3>3.5 OpenMP parallel computation</h3> -<p>SoundTouch 1.9 onwards support running the algorithms parallel in several CPU -cores. Based on benchmark the experienced multi-core processing speed-up gain -ranges between +30% (on a high-spec dual-core x86 Windows PC) to 215% (on a moderately low-spec -quad-core ARM of Raspberry Pi2). </p> -<p>See an external blog article with more detailed discussion about the -<a href="http://www.softwarecoven.com/parallel-computing-in-embedded-mobile-devices/"> -SoundTouch OpenMP optimization</a>.</p> -<p>The parallel computing support is implemented using OpenMP spec 3.0 -instructions. These instructions are supported by Visual C++ 2008 and later, and -GCC v4.2 and later. Compilers that do not supporting OpenMP will ignore these -optimizations and routines will still work properly. Possible warnings about -unknown #pragmas are related to OpenMP support and can be safely ignored.</p> -<p>The OpenMP improvements are disabled by default, and need to be enabled by -developer during compile-time. Reason for this is that parallel processing adds -moderate runtime overhead in managing the multi-threading, so it may not be -necessary nor desirable in all applications. For example real-time processing -that is not constrained by CPU power will not benefit of speed-up provided by -the parallel processing, in the contrary it may increase power consumption due -to the increased overhead.</p> -<p>However, applications that run on low-spec multi-core CPUs and may otherwise -have possibly constrained performance will benefit of the OpenMP improvements. -This include for example multi-core embedded devices.</p> -<p>OpenMP parallel computation can be enabled before compiling SoundTouch -library as follows:</p> -<ul> - <li><strong>Visual Studio</strong>: Open properties for the <strong>SoundTouch - </strong>sub-project, browse to <strong>C/C++</strong> and <strong>Language - </strong>settings. Set - there "<strong>OpenMP support</strong>" to "<strong>Yes</strong>". Alternatively add - <strong>/openmp</strong> switch to command-line - parameters</li> - <li><strong>GNU</strong>: Run the configure script with "<strong>./configure - --enable-openmp</strong>" switch, then run make as usually</li> - <li><strong>Android</strong>: Add "<strong>-fopenmp</strong>" switches to compiler & linker - options, see README-SoundTouch-Android.html in the source code package for - more detailed instructions.</li> -</ul> -<hr> -<h2><a name="SoundStretch"></a>4. SoundStretch audio processing utility -</h2> -<p>SoundStretch audio processing utility<br> - Copyright (c) Olli Parviainen 2002-2015</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 "SoundTouch" 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 width="100%" border="0" cellpadding="2"> - <tbody> - <tr> - <td valign="top"> - <pre>"infilename"</pre> - </td> - <td valign="top">Name of the input sound data file (in .WAV audio -file format). Give "stdin" as filename to use standard input pipe. </td> - </tr> - <tr> - <td valign="top"> - <pre>"outfilename"</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 don't want to save the output (e.g. when -only calculating BPM rate with '-bpm' switch). Give "stdout" as -filename to use standard output pipe.</td> - </tr> - <tr> - <td valign="top"> - <pre>[switches]</pre> - </td> - <td valign="top">Are one or more control switches.</td> - </tr> - </tbody> -</table> -<p>Available control switches are:</p> -<table width="100%" border="0" cellpadding="2"> - <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 "-tempo" switch is ignored. If "=n" is omitted, i.e. -switch "-bpm" 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 "stdin" -and "stdout" 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 "-naa" 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 <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 -"originalfile.wav" by 12.5% and stores result to file -"destinationfile.wav":</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 "orig.wav" by two semitones and stores the result to file -"dest.wav":</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 "orig.wav" 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 "orig.wav" -and adjusts the tempo to match 100 beats per minute. Result is stored -to file "dest.wav":</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> -<p><strong>Example 6</strong></p> -<p>The following command tunes song from original 440Hz tuning to 432Hz tuning: -this corresponds to lowering the pitch by -0.318 semitones:</p> -<blockquote> - <pre>soundstretch original.wav output.wav -pitch=-0.318</pre> -</blockquote> -<hr> -<h2>5. Change History</h2> -<h3>5.1. SoundTouch library Change History </h3> - <p><b>2.1.1:</b></p> - <ul> - <li>Bugfixes: Fixed potential buffer overwrite bugs in WavFile routines. Replaced asserts with runtime exceptions.</li> - <li>Android: Migrated the SoundTouch Android example to new Android Studio</li> - <li>Automake: unset ACLOCAL in bootstrap script in case earlier build script has set it</li> - - </ul> - <p><b>2.1:</b></p> - <ul> - <li>Refactored C# interface example</li> - <li>Disable anti-alias filter when switch - SOUNDTOUCH_PREVENT_CLICK_AT_RATE_CROSSOVER defined because anti-alias - filter cause slight click if the rate change crosses zero during - processing</li> - <li>Added script for building SoundTouchDll dynamic-link-library for GNU platforms</li> - <li>Rewrote Beats-per-Minute analysis algorithm for more reliable BPM detection</li> - <li>Added BPM functions to SoundTouchDll API</li> - <li>Migrated Visual Studio project files to MSVC 201x format</li> - <li>Replaced function parameter value asserts with runtime exceptions</li> - <li>Code maintenance & style cleanup</li> - </ul> - <p><b>2.0:</b></p> - <ul> - <li>Added functions to get initial processing latency, duration ratio between the original input and processed output tracks, and clarified reporting of input/output batch sizes</li> - <li>Fixed issue that added brief sequence of silence to beginning of output audio</li> - <li>Adjusted algorithm parameters to reduce reverberating effect at tempo slowdown</li> - <li>Bugfix: Fixed a glitch that could cause negative array indexing in quick seek algorithm</li> - <li>Bugfix: flush() didn't properly flush final samples from the pipeline on 2nd time in case that soundtouch object instance was recycled and used for processing a second audio stream.</li> - <li>Bugfix: Pi value had incorrect 9th/10th decimals</li> - <li>Added C# example application that uses SoundTouch dll library for processing MP3 files</li> - </ul> - <p><b>1.9.2:</b></p> - <ul> - <li>Fix in GNU package configuration</li> - </ul> - <p><b>1.9.1:</b></p> - <ul> - <li>Improved SoundTouch::flush() function so that it returns precisely the desired amount of samples for exact output duration control</li> - <li>Redesigned quickseek algorithm for improved sound quality when using the quickseek mode. The new quickseek algorithm can find 99% as good results as the - default full-scan mode, while the quickseek algorithm is remarkable less - CPU intensive.</li> - <li>Added adaptive integer divider scaling for improved sound quality when using integer processing algorithm - </li> - </ul> -<p><b>1.9:</b></p> -<ul> - <li>Added support for parallel computation support via OpenMP primitives for better performance in multicore systems. - Benchmarks show that achieved parallel processing speedup improvement - typically range from +30% (x86 dual-core) to +180% (ARM quad-core). The - OpenMP optimizations are disabled by default, see OpenMP notes above in this - readme file how to enabled these optimizations.</li> - <li>Android: Added support for Android devices featuring X86 and MIPS CPUs, - in addition to ARM CPUs.</li> - <li>Android: More versatile Android example application that processes WAV - audio files with SoundTouch library</li> - <li>Replaced Windows-like 'BOOL' types with native 'bool'</li> - <li>Changed documentation token to "dist_doc_DATA" in Makefile.am file</li> - <li>Miscellaneous small fixes and improvements</li> -</ul> -<p><b>1.8.0:</b></p> -<ul> - <li>Added support for multi-channel audio processing</li> - <li>Added support for <b>cubic</b> and <b>shannon</b> interpolation for rate and pitch shift effects besides - the original <b>linear</b> interpolation, to reduce aliasing at high frequencies due to interpolation. - Cubic interpolation is used as default for floating point processing, and linear interpolation for integer - processing.</li> - <li>Fixed bug in anti-alias filtering that limited stop-band attenuation to -10 dB instead of <-50dB, and - increased filter length from 32 to 64 taps to further reduce aliasing due to frequency folding.</li> - <li>Performance improvements in cross-correlation algorithm</li> - <li>Other bug and compatibility fixes</li> -</ul> -<p><b>1.7.1:</b></p> -<ul> - <li>Added files for Android compilation -</ul> -<p><b>1.7.0:</b></p> -<ul> - <li>Sound quality improvements/li> - <li>Improved flush() to adjust output sound stream duration to match better with - ideal duration</li> - <li>Rewrote x86 cpu feature check to resolve compatibility problems</li> - <li>Configure script automatically checks if CPU supports mmx & sse compatibility for GNU platform, and - the script support now "--enable-x86-optimizations" switch to allow disabling x86-specific optimizations.</li> - <li>Revised #define conditions for 32bit/64bit compatibility</li> - <li>gnu autoconf/automake script compatibility fixes</li> - <li>Tuned beat-per-minute detection algorithm</li> -</ul> -<p><b>1.6.0:</b></p> -<ul> - <li> Added automatic cutoff threshold adaptation to beat detection -routine to better adapt BPM calculation to different types of music</li> - <li> Retired 3DNow! optimization support as 3DNow! is nowadays -obsoleted and assembler code is nuisance to maintain</li> - <li>Retired "configure" file from source code package due to -autoconf/automake versio conflicts, so that it is from now on to be -generated by invoking "boostrap" script that uses locally available -toolchain version for generating the "configure" file</li> - <li>Resolved namespace/label naming conflicts with other libraries by -replacing global labels such as INTEGER_SAMPLES with more specific -SOUNDTOUCH_INTEGER_SAMPLES etc.<br> - </li> - <li>Updated windows build scripts & project files for Visual -Studio 2008 support</li> - <li> Updated SoundTouch.dll API for .NET compatibility</li> - <li> Added API for querying nominal processing input & output -sample batch sizes</li> -</ul> -<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: - <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.</li> - <li> Check against empty buffers in FIFOSampleBuffer</li> - <li> Other minor fixes & 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</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>1.3.1: </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>1.3.0: </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>1.2.1: </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>1.2.0: </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> |