diff options
author | Koen Vos <koen.vos@skype.net> | 2011-12-13 14:47:31 -0500 |
---|---|---|
committer | Jean-Marc Valin <jmvalin@jmvalin.ca> | 2011-12-13 14:47:31 -0500 |
commit | bf75c8ec4d0dded188bc7793de6da56c7ff0be1c (patch) | |
tree | c21bb05faf0a203b73e8f8aae90fc63dac27625d /silk/float/SigProc_FLP.h | |
parent | 6619a736376221f2782cecff55d051c3ecfc2ff7 (diff) | |
download | opus-bf75c8ec4d0dded188bc7793de6da56c7ff0be1c.tar.gz |
SILK fixes following last codec WG meeting
decoder:
- fixed incorrect scaling of filter states for the smallest quantization
step sizes
- NLSF2A now limits the prediction gain of LPC filters
encoder:
- increased damping of LTP coefficients in LTP analysis
- increased white noise fraction in noise shaping LPC analysis
- introduced maximum total prediction gain. Used by Burg's method to
exit early if prediction gain is exceeded. This improves packet
loss robustness and numerical robustness in Burg's method
- Prefiltered signal is now in int32 Q10 domain, from int16 Q0
- Increased max number of iterations in CBR gain control loop from 5 to 6
- Removed useless code from LTP scaling control
- Optimization: smarter LPC loop unrolling
- Switched default win32 compile mode to be floating-point
resampler:
- made resampler have constant delay of 0.75 ms; removed delay
compensation from silk code.
- removed obsolete table entries (~850 Bytes)
- increased downsampling filter order from 16 to 18/24/36 (depending on
frequency ratio)
- reoptimized filter coefficients
Diffstat (limited to 'silk/float/SigProc_FLP.h')
-rw-r--r-- | silk/float/SigProc_FLP.h | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/silk/float/SigProc_FLP.h b/silk/float/SigProc_FLP.h index 7b2ba089..0d70ca1b 100644 --- a/silk/float/SigProc_FLP.h +++ b/silk/float/SigProc_FLP.h @@ -50,8 +50,7 @@ void silk_bwexpander_FLP( /* compute inverse of LPC prediction gain, and */ /* test if LPC coefficients are stable (all poles within unit circle) */ /* this code is based on silk_FLP_a2k() */ -opus_int silk_LPC_inverse_pred_gain_FLP( /* O returns 1 if unstable, otherwise 0 */ - silk_float *invGain, /* O inverse prediction gain, energy domain */ +silk_float silk_LPC_inverse_pred_gain_FLP( /* O return inverse prediction gain, energy domain */ const silk_float *A, /* I prediction coefficients [order] */ opus_int32 order /* I prediction order */ ); @@ -108,9 +107,9 @@ void silk_insertion_sort_decreasing_FLP( silk_float silk_burg_modified_FLP( /* O returns residual energy */ silk_float A[], /* O prediction coefficients (length order) */ const silk_float x[], /* I input signal, length: nb_subfr*(D+L_sub) */ + const silk_float minInvGain, /* I minimum inverse prediction gain */ const opus_int subfr_length, /* I input signal subframe length (incl. D preceeding samples) */ const opus_int nb_subfr, /* I number of subframes stacked in x */ - const silk_float WhiteNoiseFrac, /* I fraction added to zero-lag autocorrelation */ const opus_int D /* I order */ ); |