Fixes packet parsing for 16-bit CPUs
[opus.git] / silk /
2018-07-22 Mark HarrisSilence compiler warnings
2018-05-24 Jean-Marc ValinFixes a SILK bandwidth switching regression
2018-05-17 Jean-Marc ValinImproving rate control for low bitrate
2018-05-14 Gustaf UllbergSilk makes use of Opus VAD
2018-03-27 Jean-Marc Valintwo more asserts I forgot
2018-03-27 Jean-Marc ValinHarden SILK x86 code
2018-03-27 Jean-Marc ValinHardening silk/fixed
2018-03-27 Jean-Marc ValinHardening asserts for SILK float
2018-03-27 Jean-Marc ValinConverting some silk_assert()s into hardening celt_asse...
2018-02-22 Gustaf UllbergFixed off-by-one issue in Silk DTX
2018-02-18 Jean-Marc ValinFixes integer overflow in SILK VAD for 10-ms frames
2017-09-28 Jean-Marc ValinBetter rate allocation for stereo SILK in hybrid mode
2017-09-11 Ralph GilesRename SSE 4.1 files to match their targets.
2017-06-13 Ralph GilesFix a duplicate symbol name.
2017-06-02 Jonathan LennoxDon't use MAY_HAVE_NEON in arm_silk_map.c.
2017-05-30 Jean-Marc Valinsilk_LIMIT_32() should return an opus_int32 (not opus_int)
2017-05-23 Linfeng Zhangcosmetics,silk: correct input/output arg comments
2017-05-17 Linfeng ZhangOptimize silk_biquad_alt_stride2() for ARM NEON
2017-05-17 Linfeng ZhangUpdate silk_biquad_alt()
2017-04-13 Linfeng ZhangOptimize silk_warped_autocorrelation_FIX() for ARM...
2017-02-27 Mark HarrisFix compiler warnings
2017-02-18 Jean-Marc ValinFix stability test
2017-02-18 Jean-Marc ValinSpeed up test_unit_LPC_inv_pred_gain by returning early...
2017-02-15 Linfeng ZhangOptimize silk_LPC_inverse_pred_gain() for ARM NEON
2017-02-15 Jean-Marc ValinAdding a unit test for LPC_inv_pred_gain()
2017-02-10 Jean-Marc ValinRemove silk_LPC_inverse_pred_gain_Q24() which is no...
2017-02-04 James Zernfix ARM build w/--disable-intrinsics --enable-asm
2017-01-30 Linfeng ZhangOptimize silk_NSQ_del_dec() for ARM NEON
2017-01-26 Jean-Marc ValinAvoids undefined behaviour in ARM-optimized code
2017-01-23 Linfeng ZhangUpdate NSQ_LPC_BUF_LENGTH macro
2017-01-18 Felicia LimAdd FIXED_POINT check if using celt_fir in SILK's LPC...
2017-01-17 Linfeng ZhangRevise celt_fir_c() to not pass in argument "mem"
2016-12-22 Mark HarrisFix some compiler warnings
2016-11-17 Mark HarrisCorrect SILK encoder gain limit
2016-11-11 Ralph GilesRemove SILK_DEBUG define when _WIN32 and _DEBUG are...
2016-11-11 Ralph GilesRemove commented-out DEBUG_STORE_DATA calls. r=keon
2016-11-02 Mark HarrisEliminate trailing spaces
2016-10-05 Jean-Marc ValinFixing some opus_int vs opus_int32 mismatches
2016-08-17 Jean-Marc ValinFixes an overflow in LPC_inverse_pred_gain_QA()
2016-08-06 Jean-Marc ValinSpeeding up PVQ using unlikely() and moving first posit...
2016-08-05 Jean-Marc ValinMaking it clearer to Coverity that nStates cannot excee...
2016-07-30 Jean-Marc ValinFixes overflow in CNG
2016-07-28 Felicia LimEnsure that NLSF cannot be negative when computing...
2016-07-25 Jean-Marc ValinDisabling the use of celt_fir() in silk_LPC_analysis_fi...
2016-07-25 Jean-Marc ValinFixes an overflow in limit_warped_coefs()
2016-07-22 Jean-Marc ValinAvoiding an overflow in the LTP quantization
2016-07-22 Jean-Marc ValinDisables MIPS silk_noise_shape_analysis_FIX() until...
2016-07-22 Jean-Marc ValinMore headroom in fixed-point silk_burg_modified() to...
2016-07-22 Jean-Marc ValinIncrease headroom in silk_warped_autocorrelation_FIX()
2016-07-22 Jean-Marc ValinIncrease headroom to avoid overflow in silk_pitch_analy...
2016-07-21 Jean-Marc ValinAvoids undefined behaviour from left-shifting negative...
2016-07-20 Tristan MatthewsFix include warning
2016-07-20 Tristan MatthewsFix include warning
2016-07-19 Jean-Marc ValinMoving FEC decision to the Opus encoder level (from...
2016-07-19 Jean-Marc ValinCBR: set gains to their previons values where we're...
2016-07-19 Jean-Marc ValinSet pulses to zero if we can't meet rate target
2016-07-19 Jean-Marc ValinNow that we're preventing huge gains, we no longer...
2016-07-19 Jean-Marc ValinCBR: lock the gain on a subframe when the number of...
2016-07-19 Koen Vosmerge back in the LTP gain safety limiter
2016-07-19 Jean-Marc ValinCBR: Using the same gain increase for fixed-point as...
2016-07-19 Koen Voscomment about reduced codelength component in LTP quant...
2016-07-19 Koen Vosavoid double if statement for finding best quantized...
2016-07-19 Koen Vosincrease NSQ decision delay from 32 to 40
2016-07-19 Koen Vosreduced noise floor in noise shaping analysis by 2 dB
2016-07-19 Koen Vosincrease order of noise shaping filter
2016-07-19 Koen Vosmore conservative scaling of LTP corrs; remove assert...
2016-07-17 Koen VosClean up: alignment of comments
2016-07-17 Koen VosClean up: replace tabs by spaces
2016-07-17 Koen Vosremoved prefilter
2016-07-17 Koen Vosslight clean up
2016-07-17 Koen Vosundo some loop unrolling
2016-07-17 Koen Vospitch analysis (in FIX) now scales the input down at...
2016-07-17 Koen Vosminor clean up of pointer offsets
2016-07-17 Koen Vossimplified computation of LTP coefs
2016-07-17 Felicia LimAttenuate SILK PLC gain only for unvoiced speech
2016-07-17 Jean-Marc ValinMakes the encoder more aggressive about meeting the...
2016-07-17 Jean-Marc ValinUsing more fine-grained complexity settings for SILK
2016-07-17 Jean-Marc ValinFixes signed integer overlof in silk_ADD_POS_SAT32()
2016-07-17 Felicia LimNew DTX that works in all modes (SILK/CELT/HYBRID)
2016-07-17 Jean-Marc ValinNLSF_VQ_MAX_SURVIVORS is no longer useful
2016-07-17 Koen Vostake advantage of more efficient NLSF quantization...
2016-07-17 Koen VosNLSF_del_dec_quant minor clean up
2016-07-17 Koen VosNLSF_encode now also uses NLSF weights from table
2016-07-17 Koen VosNLSF VQ now uses absolute error, and prediction
2016-07-17 Koen VosNLSF decoding now uses tables for NLSF weights
2016-07-17 Jean-Marc ValinImplement sum_sqr_shift() using two passes with no...
2016-07-17 Jean-Marc ValinQuality: Increase CELT rate for voiced frames in hybrid...
2016-07-15 Jean-Marc ValinApply 02f3ac225f99 fix for divide-by-zero to the SSE4...
2016-07-06 Jonathan LennoxAdd Aarch64 intrinsics for saturated add/subtract.
2016-07-06 Jonathan LennoxAdd OPUS_FAST_INT64 definition of silk_SMULWT.
2016-07-06 Jonathan LennoxExplicitly cast results of silk OPUS_FAST_INT64 macros...
2016-07-06 Jonathan LennoxMove OPUS_FAST_INT64 definition to celt/arch.h.
2016-07-06 Jonathan LennoxAdd Neon intrinsics for Silk noise shape feedback loop.
2016-07-06 Jonathan LennoxApply Neon short prediction optimization to silk_noise_...
2016-07-06 Jonathan LennoxAdd Neon intrinsics for Silk noise shape quantization.
2016-07-06 Mark HarrisUpdate URLs
2016-07-05 Jean-Marc ValinReplace another useless 32x16 multiply with 16x16
2016-07-05 Jean-Marc ValinReplacing some 32x16 multiplies with 16x16 when both...
2016-07-04 Ralph GilesAssert non-overlapping memcpy. r=mark4o
2016-06-20 Jean-Marc ValinFixes potential overflow on 16-bit platforms
next