 Jean-Marc Valin
|
625d29104b
SILK overflow fixes
|
8 years ago |
|
Jean-Marc Valin
|
9a0eb035be
More headroom in fixed-point silk_burg_modified() to avoid an overflow
|
8 years ago |
|
Jean-Marc Valin
|
3af80a93bc
Increase headroom in silk_warped_autocorrelation_FIX()
|
8 years ago |
|
Jean-Marc Valin
|
07e61f4212
Increase headroom to avoid overflow in silk_pitch_analysis_core()
|
8 years ago |
|
Jean-Marc Valin
|
ca0d38395c
PLC overflow fix
|
8 years ago |
|
Jean-Marc Valin
|
c952f90b4f
Prevent overflows in PLC celt_iir()
|
8 years ago |
|
Jean-Marc Valin
|
1fba9045b1
fixes overflows in celt_iir() by doing proper saturation
|
8 years ago |
|
Jean-Marc Valin
|
21036d062f
getting rid of a PLC overflow (I think)
|
8 years ago |
|
Jean-Marc Valin
|
33f2d6c18d
post-filter saturation
|
8 years ago |
|
Jean-Marc Valin
|
d99c71364b
saturate MDCT output
|
8 years ago |
|
Jean-Marc Valin
|
273906404c
silencing overflows in MDCT and FFT
|
8 years ago |
|
Jean-Marc Valin
|
ae33218495
Reduce risk of overflowing in mono downsampling
|
8 years ago |
|
Jean-Marc Valin
|
c76c4de123
Avoid overflowing on ridiculously large energy
|
8 years ago |
|
Jean-Marc Valin
|
b290bca91a
Fixes shift<-2 case for denormalise_bands()
|
8 years ago |
 Mark Harris
|
43b47a41b1
Silence fixed point unused parameter warning
|
8 years ago |
|
Mark Harris
|
b0b6d6aeac
test_opus_decode: Check all soft clip samples
|
8 years ago |
|
Mark Harris
|
d401961a18
tests: Avoid shift into sign undefined behavior
|
8 years ago |
|
Jean-Marc Valin
|
de473b305c
Avoids undefined behaviour from left-shifting negative values
|
8 years ago |
 Felicia Lim
|
e1bce7b3f6
Fix use_dtx for DISABLE_FLOAT_API
|
8 years ago |
 Tristan Matthews
|
6f26011129
Fix include warning
|
8 years ago |
|
Tristan Matthews
|
c4027c1d34
Fix include warning
|
8 years ago |
|
Jean-Marc Valin
|
6fccb4b615
Adds some smoothing to the energy quantization
|
8 years ago |
|
Jean-Marc Valin
|
b3c9e79a82
Tune SILK rate for hybrid with FEC enabled
|
8 years ago |
|
Jean-Marc Valin
|
2b011b64d8
Moving FEC decision to the Opus encoder level (from SILK level)
|
8 years ago |
|
Jean-Marc Valin
|
099fc05ebf
Taking loss rate into account in compute_equiv_rate()
|
8 years ago |
|
Jean-Marc Valin
|
7780d4a6b1
Retuning the tf_analysis() lambda
|
8 years ago |
|
Jean-Marc Valin
|
1484591656
Fixes equiv_rate for CBR
|
8 years ago |
|
Jean-Marc Valin
|
2ca6df03df
Minor TF cleanup (tf_sum was useless), plus comments
|
8 years ago |
|
Jean-Marc Valin
|
f705e9b5bd
Use SPREAD_AGGRESSIVE on non-transient hybrid frames
|
8 years ago |
|
Jean-Marc Valin
|
8229f07d3b
Don't use too much temporal resolution on hybrid frames at low rate
|
8 years ago |
