jm
|
afaa98ad53
lpc floor converted to fixed-point
|
19 years ago |
jm
|
5144638319
Improved packet loss concealment
|
19 years ago |
jm
|
3ec0d16569
removed unused buffer
|
19 years ago |
jm
|
f5228ae08e
Moved all the includes from /usr/include to /usr/include/speex
|
20 years ago |
jm
|
5f5971d6fa
speex_encode/speex_decode are back to using floats, new speex_encode_int
|
20 years ago |
jm
|
556c879aae
Added const's all over the place
|
21 years ago |
jm
|
384dd357e1
fixed-point: bw_lpc and lpc_to_lsp are now done.
|
21 years ago |
jm
|
c42830cfe3
fixed-point: integerized pi_gain's
|
21 years ago |
jm
|
b3ef4f2621
cleanup
|
21 years ago |
jm
|
c19c2acc00
fixed-point: LSPs are now stored quantized
|
21 years ago |
jm
|
766ab3fb94
fixed-point: QMF entirely in fixed-point now
|
21 years ago |
jm
|
b1004ff2a9
fixed-point: converted user-visible functions to use "short" signals,
|
21 years ago |
jm
|
86a06fff22
fixed-point: converted all signals to spx_sig_t
|
21 years ago |
jm
|
91e19372fa
fixed-point: more signal scaling again, some auto-correlation work
|
21 years ago |
jm
|
3a6c1b54ca
fixed-point: more lpc stuff
|
21 years ago |
jm
|
b958f56414
first step in fixed-point port, converted the LPC filters
|
21 years ago |
jm
|
6bb8afa4df
added a call telling the codec not to encode the submode in the frame.
|
21 years ago |
jm
|
bdff410d46
Minor cleanup (who needs reflection coefficients anyway) in LPC code.
|
21 years ago |
jm
|
067ddb369b
Added a return value (error) to the *ctl functions, added re-allocation
|
22 years ago |
jm
|
c511551c96
Cleaned up mem allocation in sb_celp.c like in nb_celp.c
|
22 years ago |
jm
|
c2a25cac3f
misc stuff for beta4
|
22 years ago |
jm
|
9b324f61be
ABR seems to work for wideband too...
|
22 years ago |
jm
|
6c8547685f
VAD should now work on wideband too.
|
22 years ago |
jm
|
9aa02be42e
First shot at high-band perceptual enhancement
|
22 years ago |
jm
|
954bdcb812
Ultra-wideband VBR seems to work. Also, fixed a bug for wideband VBR.
|
22 years ago |
jm
|
7e23305184
Fixed a bug in the VBR analyzer, trying to re-tune it... also, started
|
22 years ago |
jm
|
45c3a62011
Update for non-standard (not 8,16,32 kHz) sampling rates, changed package
|
22 years ago |
jm
|
f69b8d8563
Wideband code cleanup, first shot at an "ultra-wideband" mode at 32 kHz,
|
22 years ago |
jm
|
5d10b870fe
The temp stack is now void* instead of float*
|
22 years ago |
jm
|
bbce5d1ba0
QMF optimizations by segher
|
22 years ago |