Implements hard CBR for SILK
This is achieved by running the encoding process in a loop and padding when we don't reach the exact rate. It also implements VBR-with-cap, which means we no longer need to artificially decrease the SILK bandwidth when it's close to the cap.
Showing
- silk/API.h 2 additions, 2 deletionssilk/API.h
- silk/SigProc_FIX.h 1 addition, 1 deletionsilk/SigProc_FIX.h
- silk/biquad_alt.c 3 additions, 3 deletionssilk/biquad_alt.c
- silk/control.h 3 additions, 0 deletionssilk/control.h
- silk/control_codec.c 1 addition, 1 deletionsilk/control_codec.c
- silk/enc_API.c 32 additions, 7 deletionssilk/enc_API.c
- silk/fixed/LTP_scale_ctrl_FIX.c 0 additions, 1 deletionsilk/fixed/LTP_scale_ctrl_FIX.c
- silk/fixed/encode_frame_FIX.c 136 additions, 30 deletionssilk/fixed/encode_frame_FIX.c
- silk/fixed/find_pred_coefs_FIX.c 1 addition, 1 deletionsilk/fixed/find_pred_coefs_FIX.c
- silk/fixed/main_FIX.h 3 additions, 1 deletionsilk/fixed/main_FIX.h
- silk/fixed/process_gains_FIX.c 4 additions, 0 deletionssilk/fixed/process_gains_FIX.c
- silk/fixed/structs_FIX.h 3 additions, 0 deletionssilk/fixed/structs_FIX.h
- silk/float/encode_frame_FLP.c 137 additions, 24 deletionssilk/float/encode_frame_FLP.c
- silk/float/find_pred_coefs_FLP.c 1 addition, 1 deletionsilk/float/find_pred_coefs_FLP.c
- silk/float/inner_product_FLP.c 6 additions, 6 deletionssilk/float/inner_product_FLP.c
- silk/float/main_FLP.h 3 additions, 1 deletionsilk/float/main_FLP.h
- silk/float/pitch_analysis_core_FLP.c 3 additions, 3 deletionssilk/float/pitch_analysis_core_FLP.c
- silk/float/process_gains_FLP.c 4 additions, 0 deletionssilk/float/process_gains_FLP.c
- silk/float/silk_float.vcxproj 1 addition, 1 deletionsilk/float/silk_float.vcxproj
- silk/float/structs_FLP.h 4 additions, 0 deletionssilk/float/structs_FLP.h
Loading
Please register or sign in to comment