Tue, 14 Jan 2014 21:37:53 +0000 |
Jamie Bullock |
Add .travis.yml
|
Mon, 06 Jan 2014 16:41:42 +0000 |
Jamie Bullock |
Update gitignore to include python byte code
|
Tue, 05 Nov 2013 17:52:38 +0000 |
Jamie Bullock |
Ignore files auto-generated by SWIG
|
Tue, 05 Nov 2013 17:52:21 +0000 |
Jamie Bullock |
Remove unnecessary "-lxtract" from python SWIG build file. See #40
|
Tue, 16 Jul 2013 20:52:45 +0100 |
Jamie Bullock |
Add noise test source to simpletest. Add xtract_flux() example to simpletest
|
Mon, 08 Jul 2013 10:14:42 +0100 |
Jamie Bullock |
Optimise xtract_smoothness() by removing call to malloc(). Also fix bug in xtract_smoothness() where *result was uninitialised but used in calculation.
|
Mon, 08 Jul 2013 09:52:19 +0100 |
Jamie Bullock |
Remove spurious argv specification from xtract_smoothness() documentation. Fixes #27.
|
Thu, 04 Jul 2013 08:59:01 +0100 |
Jamie Bullock |
Use log10 (double precision) instead of log10f. Fixes bug in spectral flatness
|
Mon, 01 Jul 2013 17:25:21 +0100 |
Jamie Bullock |
Remove spurious print statement
|
Wed, 19 Jun 2013 16:22:47 -0700 |
Jamie Bullock |
Offset peak interpolation by one spectral bin. Fixes #20
|
Wed, 19 Jun 2013 16:05:59 -0700 |
Jamie Bullock |
Uncomment free(), fixing memory leak
|
Wed, 19 Jun 2013 16:04:42 -0700 |
Jamie Bullock |
Don't read past end of spectral data array
|
Wed, 19 Jun 2013 15:54:52 -0700 |
Jamie Bullock |
Remove extraneous print statement
|
Wed, 19 Jun 2013 15:30:04 -0700 |
Jamie Bullock |
Initialise ooura struct with right amount of memory for ooura FFT. Fixes #11
|
Wed, 19 Jun 2013 15:15:21 -0700 |
Jamie Bullock |
Don't include default CFLAGS in configure
|
Wed, 19 Jun 2013 14:59:58 -0700 |
Jamie Bullock |
Fix memory leak by calling xtract_free_window()
|
Wed, 19 Jun 2013 17:10:22 +0100 |
Jamie Bullock |
Print which FFT implementation is being used at startup
|
Tue, 18 Jun 2013 13:23:17 -0700 |
Jamie Bullock |
Dynamically allocate storage in simpltest example. Ooura doesn't seem to like the being passed memory on the stack on Linux. Fixes #11
|
Tue, 18 Jun 2013 13:21:02 -0700 |
Jamie Bullock |
Copy out input values for Ooura so the in-place transform doesn't overwrite *data
|
Tue, 18 Jun 2013 08:51:28 -0700 |
Jamie Bullock |
Update definition with const to reflect prototype
|
Tue, 18 Jun 2013 08:17:05 -0700 |
Jamie Bullock |
Comment out unused variable
|
Tue, 18 Jun 2013 08:16:45 -0700 |
Jamie Bullock |
Update function prototype to use const for passed in sample buffer
|
Tue, 18 Jun 2013 08:15:43 -0700 |
Jamie Bullock |
Increment feature count to correct value
|
Tue, 18 Jun 2013 08:10:23 -0700 |
Jamie Bullock |
Define M_PI to 3.14159265358979323846264338327 on platforms where it is not provided by math.h
|
Sun, 02 Jun 2013 22:55:18 +0100 |
Jamie Bullock |
More xtract functions included in simpletest example
|
Sun, 02 Jun 2013 22:54:41 +0100 |
Jamie Bullock |
Pass block size to xtract_init_vdsp_data() instead of block size / 2. Fixes bug with erroneous values in second half of spectrum when using vDSP FFT
|
Fri, 31 May 2013 23:01:59 +0100 |
Jamie Bullock |
Reinstate path fix
|
Fri, 31 May 2013 22:57:13 +0100 |
Jamie Bullock |
Remove obsolete reference to xtract_scalar_private.h
|
Fri, 31 May 2013 22:44:03 +0100 |
Jamie Bullock |
Add wavelet-based pitch tracker
|
Fri, 31 May 2013 22:43:17 +0100 |
Jamie Bullock |
Add .gitignore
|
Wed, 24 Apr 2013 15:10:51 +0100 |
Jamie Bullock |
Provide relative include path for libxtract.h for compatibility with non-autotools builds
|
Fri, 15 Mar 2013 05:46:57 -0700 |
Jamie Bullock |
Merge pull request #13 from felixonmars/master
|
Fri, 08 Feb 2013 19:50:27 +0800 |
Felix Yan |
fix for automake 1.13
|
Fri, 15 Mar 2013 10:29:08 +0000 |
Jamie Bullock |
Remove erroneous casts to void * in xtract() function calls
|
Thu, 14 Mar 2013 13:03:12 +0000 |
Jamie Bullock |
Add explicit casts to malloc() calls for C++ compatibility
|
Thu, 10 Jan 2013 22:41:03 +0000 |
Jamie Bullock |
fixed Linux bugs
|
Thu, 10 Jan 2013 21:43:57 +0000 |
Jamie Bullock |
added LICENSE
|
Wed, 09 Jan 2013 23:22:07 +0000 |
Jamie Bullock |
removed extraneous files
|
Wed, 09 Jan 2013 23:18:25 +0000 |
Jamie Bullock |
removed references to GPL from README
|
Wed, 09 Jan 2013 23:09:34 +0000 |
Jamie Bullock |
implemented optimised FFT via the Accelerate framework. closes #5
|
Wed, 09 Jan 2013 20:15:50 +0000 |
Jamie Bullock |
add option to explicitly specify Ooura library at configure time
|
Wed, 09 Jan 2013 15:59:48 +0000 |
Jamie Bullock |
fixed bug in xtract_dct where data wasn't being copied to result vector
|
Wed, 09 Jan 2013 12:54:28 +0000 |
Jamie Bullock |
run FFT in place on the data vector
|
Wed, 09 Jan 2013 12:45:29 +0000 |
Jamie Bullock |
switched from single to double precision througout. closes #9
|
Tue, 08 Jan 2013 16:44:41 +0000 |
Jamie Bullock |
fixed build system bugs to do with the maintainer-clean target
|
Tue, 08 Jan 2013 14:32:45 +0000 |
Jamie Bullock |
removed floatArray and intArray from Java SWIG bindings
|
Mon, 07 Jan 2013 18:14:34 +0000 |
Jamie Bullock |
removed references to FFTW
|
Mon, 07 Jan 2013 17:03:10 +0000 |
Jamie Bullock |
Merge branch 'master' of https://github.com/jamiebullock/LibXtract
|
Mon, 07 Jan 2013 17:01:14 +0000 |
Jamie Bullock |
changed license from GPL to MIT. closes #6
|
Mon, 07 Jan 2013 16:27:15 +0000 |
Jamie Bullock |
added Ooura implementation to repository
|
Tue, 11 Dec 2012 02:58:18 -0800 |
Jamie Bullock |
Merge pull request #4 from tchajed/master
|
Sun, 02 Dec 2012 13:51:23 -0600 |
Tej Chajed |
removing unneeded %newobject decl
|
Sun, 02 Dec 2012 13:50:46 -0600 |
Tej Chajed |
adding ctype.i include
|
Thu, 13 Sep 2012 10:56:43 +0100 |
Jamie Bullock |
Changed references to SVN in README.md to GIT
|
Wed, 05 Sep 2012 16:27:18 +0100 |
Jamie Bullock |
Updated simpletest example to include MFCC
|
Mon, 18 Jun 2012 21:38:14 +0100 |
Jamie Bullock |
SWIG Java bindings improvements
|
Wed, 13 Jun 2012 15:08:11 +0100 |
Jamie Bullock |
Java SWIG bindings improvements
|
Sun, 10 Jun 2012 12:11:15 +0100 |
Jamie Bullock |
Updated README to cater for release vs SVN
|
Thu, 31 May 2012 19:48:50 +0100 |
Jamie Bullock |
removed PPC support
|
Thu, 31 May 2012 16:27:57 +0100 |
Jamie Bullock |
renamed README to README.md
|
Thu, 31 May 2012 16:25:17 +0100 |
Jamie Bullock |
converted README to Markdown
|
Fri, 30 Mar 2012 13:26:41 +0000 |
Jamie Bullock |
- minor SWIG fix
|
Fri, 30 Mar 2012 13:13:52 +0000 |
Jamie Bullock |
- updated python test to import libxtract.xtract
|
Thu, 29 Mar 2012 17:24:45 +0000 |
Jamie Bullock |
- updated to latest SWIG macros
|
Thu, 29 Mar 2012 12:23:04 +0000 |
Jamie Bullock |
- fixed build-time bug. in some cases average_deviation was commented out instead of spectral_average_deviation
|
Thu, 31 Mar 2011 12:22:29 +0000 |
Jamie Bullock |
- fixed xtract_spread() -- now just calls xtract_spectral_variance()
|
Wed, 30 Mar 2011 10:51:39 +0000 |
Jamie Bullock |
- reimplemented xtract_spectral_variance() xtract_spectral_skewness() and xtract_spectral_kurtosis() using correct maths
|
Wed, 30 Mar 2011 10:05:07 +0000 |
Jamie Bullock |
- fixed typos in *result initialisation potentially fixing horrible bug
|
Mon, 23 Aug 2010 11:28:39 +0000 |
Jamie Bullock |
- Max example tweaks
|
Wed, 03 Feb 2010 22:35:13 +0000 |
Jamie Bullock |
- fixed DC/Nyquist inclusion bug in xtract_spectrum() and refactored a bit
|
Tue, 11 Nov 2008 11:55:55 +0000 |
Jamie Bullock |
- Committed patch by Chris Cannam to avoid reading past the end of an array in peak detection routine in xtract_peak_spectrum()
|
Fri, 12 Sep 2008 13:52:20 +0000 |
Jamie Bullock |
- Committing ancient change left on HD. Relates to FFTW fft type
|
Sun, 04 May 2008 11:02:40 +0000 |
Jamie Bullock |
Fixed bug in peak interpolation algorithm in xtract_peak_spectrum()
|
Wed, 26 Mar 2008 13:04:33 +0000 |
Jamie Bullock |
- Fixed build fail if --enable-fft not specified
|
Sat, 16 Feb 2008 20:13:05 +0000 |
Jamie Bullock |
- Added to pd example the ability to differentiate between different
|
Fri, 15 Feb 2008 15:49:49 +0000 |
Jamie Bullock |
Added new extraction function: xtract_subbands()
|
Fri, 15 Feb 2008 12:43:13 +0000 |
Jamie Bullock |
- Fixed bugs in xtract_flatness(), or at least added necessary
|
Wed, 02 Jan 2008 04:10:21 +0000 |
Jamie Bullock |
- Minor fix in xtract_lpc... or at least think it's a fix
|
Wed, 02 Jan 2008 02:26:13 +0000 |
Jamie Bullock |
- Re-factoring in xtract_spectrum and fixed normalisation bug
|
Tue, 01 Jan 2008 16:17:44 +0000 |
Jamie Bullock |
- Improvements to SWIG bindings generation script
|
Sat, 29 Dec 2007 18:35:35 +0000 |
Jamie Bullock |
- added bark coefficients PD example
|
Sat, 29 Dec 2007 17:33:17 +0000 |
Jamie Bullock |
- PD example brought in line with new delta features and subframe function
|
Fri, 28 Dec 2007 19:34:51 +0000 |
Jamie Bullock |
- Added new helper functions: xtract_windowed() and xtract_features_from_subframes()
|
Thu, 27 Dec 2007 20:37:15 +0000 |
Jamie Bullock |
Added l-norm/spectral flux
|
Thu, 27 Dec 2007 17:51:07 +0000 |
Jamie Bullock |
- Added extra argument to xtract_spectrum to give the option of normalising the magnitude/power coeffificients
|
Mon, 24 Dec 2007 13:21:13 +0000 |
Jamie Bullock |
- Fixes to descriptors.c where no break statement was given for certain cases is switch conditionals
|
Fri, 21 Dec 2007 11:05:20 +0000 |
Jamie Bullock |
- Added some previously uncommitted Pd testing examples
|
Tue, 16 Oct 2007 09:37:06 +0000 |
Jamie Bullock |
Fix for 'multiple symbol definitions' vs 'symbol not defined' dilemna. I think the solution is to wrap the globals in a struct, declare it in a header, and wrap with a definition guard, then define _once_ at library init time. (Sounds like a recipe for something...)
|
Tue, 09 Oct 2007 16:33:06 +0000 |
Jamie Bullock |
Added missing extern storage class specifier to private globals
|
Tue, 09 Oct 2007 13:47:58 +0000 |
Dan Stowell |
Fixed bug in xtract_init_mfcc() - iteration to decide filter peak positions only filled in freq_bands+1 values, when it should fill in freq_bands+2
|
Mon, 08 Oct 2007 10:32:16 +0000 |
Jamie Bullock |
Committed missing files needed for new out-of-place fftw plan creation
|
Sat, 06 Oct 2007 16:36:00 +0000 |
Jamie Bullock |
Removed fftw_plan from xtraction functions. Created new init function xtract_init_fft() for creating plans, which have global scope. Updated examples to reflect the change. New configure option: --with-fft_optimisation (0 = FFTW_ESTIMATE, 1 = FFTW_MEASURE, 2 = FFTW_PATIENT).
|
Thu, 04 Oct 2007 14:29:30 +0000 |
Jamie Bullock |
Incremented FIX revision number and updated ChangeLog
|
Wed, 03 Oct 2007 13:43:16 +0000 |
Dan Stowell |
Dan Stowell: Removed strange "zeroing" part of xtract_mfcc() which was zeroing a load of elements despite the fact that they're ignored by the DCT process called next, and never used for anything. This was writing to an assumed large result array (same size as number of FFT bins) despite the fact that only a small number of MFCCs (typically less than 50) are required, therefore either wasting memory or writing to memory it shouldn't do!
|
Wed, 03 Oct 2007 10:01:15 +0000 |
Dan Stowell |
Minor modifications to MFCC init, mainly just code formatting
|
Mon, 24 Sep 2007 10:26:26 +0000 |
Jamie Bullock |
Minor documentation change
|