aboutsummaryrefslogtreecommitdiff
path: root/celt
AgeCommit message (Expand)Author
2017-06-06Optimize fixed-point celt_inner_prod() and dual_inner_prod() for ARM NEONLinfeng Zhang
2017-06-06Replace call of celt_inner_prod_c() (step 2)Linfeng Zhang
2017-06-06Replace call of celt_inner_prod_c() (step 1)Linfeng Zhang
2017-06-02Recalibrate VBR to hit target bitrate on averageJean-Marc Valin
2017-06-02Update dynalloc VBR calibrationJean-Marc Valin
2017-06-01Adding leakage modelling to boost bandsJean-Marc Valin
2017-06-01Let CBR use more bits for dynallocJean-Marc Valin
2017-05-25Fixes unit tests that need arch-specific codeJean-Marc Valin
2017-05-24Fixes a double->float conversion warningJean-Marc Valin
2017-05-24Fixes some fixed-point 16-bit int overflowsJean-Marc Valin
2017-05-24Reducing trim at low bitrateJean-Marc Valin
2017-02-26Fix compiler warningsMark Harris
2017-02-18Fix sum initializer in celt_fir()Jean-Marc Valin
2017-02-17Remove SMALL_FOOTPRINT code for celt_fir()Jean-Marc Valin
2017-02-15Increasing GLOBAL_STACK_SIZE to 120000 to avoid failureJean-Marc Valin
2017-02-15Reduce the scope of Ne10 includesMichael Bradshaw
2017-02-15Remove #include "config.h" from ARM Ne10 headersMichael Bradshaw
2017-02-04arm2gnu.pl: Fix GNU ARM .type directiveMark Harris
2017-01-25Avoids undefined behaviour in ARM-optimized codeJean-Marc Valin
2017-01-17Revise celt_fir_c() to not pass in argument "mem"Linfeng Zhang
2017-01-12fixed_debug.h: delete some dead codeJames Zern
2017-01-05Avoids pre-echo in hybrid mode caused by noise being injected in the first bandJean-Marc Valin
2016-12-22Fix some compiler warningsMark Harris
2016-12-20Makes analysis run at 24 kHz, with 20-ms framesJean-Marc Valin
2016-11-04Removes OPUS_FRAMESIZE_VARIABLEJean-Marc Valin
2016-11-02Improves CELT bit allocation at very low bitratesJean-Marc Valin
2016-10-29Don't use hybrid "weak transients" on vowelsJean-Marc Valin
2016-10-28Define "weak transients" for low bitrate hybridJean-Marc Valin
2016-10-27Fixes an unstable energy issue for low-bitrate hybridJean-Marc Valin
2016-09-14Prevents ridiculously large gains from causing inf/NaNs in float decoderJean-Marc Valin
2016-09-01Make it possible to ignore inverted phase stereo for downmix purposesJean-Marc Valin
2016-09-01Bitstream change: prevent noise injection in hybrid modeJean-Marc Valin
2016-08-15Removed a float operation that sneaked in in the fixed-point codeJean-Marc Valin
2016-08-15Speeding up PVQ search by allocating even more pulses in the projection.Jean-Marc Valin
2016-08-14Apply deemphasis to both channels in the same loop for the simple caseJean-Marc Valin
2016-08-13Making gcc use SSE directly for float->int conversion when availableJean-Marc Valin
2016-08-12Reducing dependencies in deemphasis()Jean-Marc Valin
2016-08-11Fixes problem with M_PIJean-Marc Valin
2016-08-11Making stereo_itheta() use the same atan2() approximation as tonality_analysis()Jean-Marc Valin
2016-08-09SSE2 implementation of the PVQ searchJean-Marc Valin
2016-08-06Making signx[] an int in alg_quant() and removes unnecessary sign copyingJean-Marc Valin
2016-08-06Speeding up PVQ using unlikely() and moving first position out of the loopJean-Marc Valin
2016-08-06Getting gcc to use cmovs rather than branches in alg_quant()Jean-Marc Valin
2016-08-05biasing quantizationJean-Marc Valin
2016-08-05Fixes "mixed declarations and code"Jean-Marc Valin
2016-08-05Weighting theta rdo based on channel energyJean-Marc Valin
2016-07-28Don't do theta RDO on intensity-stereo-coded bandsJean-Marc Valin
2016-07-28Saving the state rather than re-compute the best optionJean-Marc Valin
2016-07-28Making stereo theta decision based on minimizing distortionJean-Marc Valin
2016-07-28controlling roundingJean-Marc Valin
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-22 00:41:52 +0000