github.com/xiph/speex.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Expand)	Author
2015-04-01	remove trailing whitespace	Tristan Matthews
2009-01-30	Cosmetic fixes	Alfred E. Heggestad
2008-08-13	Added an option to implement iir_mem16() and fir_mem16() in terms of	Jean-Marc Valin
2008-08-06	Making it possible to disable either the encoder or the decoder using macros	Jean-Marc Valin
2007-11-08	Moved the remaining of misc.h into arch.h	jm
2007-10-30	Patch by Thom Johansen: change memory of qmf_synth to 16-bit.	jm
2007-03-16	Doing some sanity checking (just in case) on the propagated long-term	jm
2006-12-08	Removed implementation of all the 32-bit filters now that everything is	jm
2006-12-08	Now using only the 16-bit version of the filters.	jm
2006-12-08	Making use of the wideband encoder input buffer to save memory: 640 bytes off	jm
2006-12-04	Removed 5 kB of useless buffering in the wideband encoder. Also, fixed the	jm
2006-12-03	Another 1 kB off the encoder state by converting the qmf_decomp output to	jm
2006-12-01	Stopped storing zeros in qmf_resynth -- saved ~800 bytes of stack memory in	jm
2006-11-29	Saved another 2kB from the decoder by converting high-band synthesis to	jm
2006-11-28	Replaced fir_mem_up by qmf_synth which doesn't require intermediate	jm
2006-11-26	More debug info and fixed two unlikely overflows on clipped input.	jm
2006-11-26	Removed the generic PSHR/SHL/SHR operators and changed them to either the	jm
2006-11-19	should no longer complain about unused var with the float build.	jm
2006-11-19	Fixed an overflow in the excitation decoding (again on heavily-clipped signal)	jm
2006-11-17	Enhancer should no longer overflow on insanely loud signals.	jm
2006-11-14	Fixed two fixed-point overflows on heavily-clipped input	jm
2006-10-20	making sure all global variables are const (for Symbian users)	jm
2006-09-08	Adding the "f" suffix to float constants (path by Michael Jerris).	jm
2006-08-04	pseudo-IRS filter (not used yet)	jm
2006-08-02	cleaning up the IO filters	jm
2006-07-30	fixed-point version of the high-pass seems to work now.	jm
2006-07-30	before I get hatemail from windows users...	jm
2006-07-30	input/output high-pass filters	jm
2006-07-29	copyright details, more info about the mdf learning rate	jm
2006-06-25	Removed old enhancer code (was already disabled)	jm
2006-06-07	Patch by David Rowe: sending the stack parameter to *_mem16() filters.	jm
2006-05-20	excitation and target are now 16-bit (were 32) in the encoder	jm
2006-05-16	removed a bunch of overflow cases, most of which were in wideband	jm
2006-05-16	decoder excitation now in 16-bit precision (was 32), which saves quite a bit	jm
2006-05-10	C89 fixes	jm
2006-05-08	cleaned up temp allocation in enhancer a bit	jm
2006-05-08	cleaned up a variable-size array (before I get flamed by angry Win32 users)	jm
2006-05-08	removed an "if" from the enhancer inner loop	jm
2006-05-08	Enable the new enhancer by default.	jm
2006-05-07	fixed-point accuracy improvements in both compute_rms16 and fir_mem_up	jm
2006-04-30	Approximation for pitch search in new enhancer. Trying to fix PLC as well.	jm
2006-04-30	cleaning up new enhancer	jm
2006-04-29	fir_mem16 implemented using filter_mem16 on Blackfin	jm
2006-04-29	blackfin assembly for iir_mem16	jm
2006-04-29	Blackfin assembly version of filter_mem16	jm
2006-04-27	new enhancer now fully fixed-point	jm
2006-04-27	two more variables in fixed-point. Oh, and un-b0rked the float again.	jm
2006-04-27	more fixed-point in new enhancer	jm
2006-04-27	three more vars in fixed-point (four to go).	jm
2006-04-26	oops. Previous checking broke the floating point. Think I fixed it.	jm