AMRContext Struct Reference


Data Fields

AVFrame avframe
 AVFrame for decoded samples.
AMRNBFrame frame
 decoded AMR parameters (lsf coefficients, codebook indexes, etc)
uint8_t bad_frame_indicator
 bad frame ? 1 : 0
enum Mode cur_frame_mode
int16_t prev_lsf_r [LP_FILTER_ORDER]
 residual LSF vector from previous subframe
double lsp [4][LP_FILTER_ORDER]
 lsp vectors from current frame
double prev_lsp_sub4 [LP_FILTER_ORDER]
 lsp vector for the 4th subframe of the previous frame
float lsf_q [4][LP_FILTER_ORDER]
 Interpolated LSF vector for fixed gain smoothing.
float lsf_avg [LP_FILTER_ORDER]
 vector of averaged lsf vector
float lpc [4][LP_FILTER_ORDER]
 lpc coefficient vectors for 4 subframes
uint8_t pitch_lag_int
 integer part of pitch lag from current subframe
float excitation_buf [PITCH_DELAY_MAX+LP_FILTER_ORDER+1+AMR_SUBFRAME_SIZE]
 current excitation and all necessary excitation history
float * excitation
 pointer to the current excitation vector in excitation_buf
float pitch_vector [AMR_SUBFRAME_SIZE]
 adaptive code book (pitch) vector
float fixed_vector [AMR_SUBFRAME_SIZE]
 algebraic codebook (fixed) vector (must be kept zero between frames)
float prediction_error [4]
 quantified prediction errors {20log10(^gamma_gc)} for previous four subframes
float pitch_gain [5]
 quantified pitch gains for the current and previous four subframes
float fixed_gain [5]
 quantified fixed gains for the current and previous four subframes
float beta
 previous pitch_gain, bounded by [0.0,SHARP_MAX]
uint8_t diff_count
 the number of subframes for which diff has been above 0.65
uint8_t hang_count
 the number of subframes since a hangover period started
float prev_sparse_fixed_gain
 previous fixed gain; used by anti-sparseness processing to determine "onset"
uint8_t prev_ir_filter_nr
 previous impulse response filter "impNr": 0 - strong, 1 - medium, 2 - none
uint8_t ir_filter_onset
 flag for impulse response filter strength
float postfilter_mem [10]
 previous intermediate values in the formant filter
float tilt_mem
 previous input to tilt compensation filter
float postfilter_agc
 previous factor used for adaptive gain control
float high_pass_mem [2]
 previous intermediate values in the high-pass filter
float samples_in [LP_FILTER_ORDER+AMR_SUBFRAME_SIZE]
 floating point samples

Detailed Description

Definition at line 97 of file amrnbdec.c.


Field Documentation

AVFrame for decoded samples.

Definition at line 98 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), and amrnb_decode_init().

bad frame ? 1 : 0

Definition at line 100 of file amrnbdec.c.

Referenced by unpack_bitstream().

previous pitch_gain, bounded by [0.0,SHARP_MAX]

Definition at line 124 of file amrnbdec.c.

Referenced by pitch_sharpening().

the number of subframes for which diff has been above 0.65

Definition at line 125 of file amrnbdec.c.

Referenced by fixed_gain_smooth().

pointer to the current excitation vector in excitation_buf

Definition at line 115 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), amrnb_decode_init(), and decode_pitch_vector().

float AMRContext::excitation_buf[PITCH_DELAY_MAX+LP_FILTER_ORDER+1+AMR_SUBFRAME_SIZE]

current excitation and all necessary excitation history

Definition at line 114 of file amrnbdec.c.

Referenced by amrnb_decode_init(), and update_state().

quantified fixed gains for the current and previous four subframes

Definition at line 122 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), fixed_gain_smooth(), and update_state().

float AMRContext::fixed_vector[AMR_SUBFRAME_SIZE]

algebraic codebook (fixed) vector (must be kept zero between frames)

Definition at line 118 of file amrnbdec.c.

Referenced by amrnb_decode_frame().

decoded AMR parameters (lsf coefficients, codebook indexes, etc)

Definition at line 99 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), decode_gains(), lsf2lsp_3(), lsf2lsp_5(), and unpack_bitstream().

the number of subframes since a hangover period started

Definition at line 126 of file amrnbdec.c.

Referenced by fixed_gain_smooth().

previous intermediate values in the high-pass filter

Definition at line 135 of file amrnbdec.c.

Referenced by amrnb_decode_frame().

flag for impulse response filter strength

Definition at line 130 of file amrnbdec.c.

Referenced by anti_sparseness().

float AMRContext::lpc[4][LP_FILTER_ORDER]

lpc coefficient vectors for 4 subframes

Definition at line 110 of file amrnbdec.c.

Referenced by amrnb_decode_frame().

float AMRContext::lsf_avg[LP_FILTER_ORDER]

vector of averaged lsf vector

Definition at line 108 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), and amrnb_decode_init().

float AMRContext::lsf_q[4][LP_FILTER_ORDER]

Interpolated LSF vector for fixed gain smoothing.

Definition at line 107 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), amrnb_decode_init(), lsf2lsp_3(), and lsf2lsp_for_mode12k2().

double AMRContext::lsp[4][LP_FILTER_ORDER]

lsp vectors from current frame

Definition at line 104 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), lsf2lsp_3(), lsf2lsp_5(), and update_state().

quantified pitch gains for the current and previous four subframes

Definition at line 121 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), anti_sparseness(), decode_gains(), pitch_sharpening(), synthesis(), and update_state().

integer part of pitch lag from current subframe

Definition at line 112 of file amrnbdec.c.

Referenced by decode_pitch_vector(), and pitch_sharpening().

float AMRContext::pitch_vector[AMR_SUBFRAME_SIZE]

adaptive code book (pitch) vector

Definition at line 117 of file amrnbdec.c.

Referenced by decode_pitch_vector(), and synthesis().

previous factor used for adaptive gain control

Definition at line 134 of file amrnbdec.c.

Referenced by postfilter().

previous intermediate values in the formant filter

Definition at line 132 of file amrnbdec.c.

Referenced by postfilter().

quantified prediction errors {20log10(^gamma_gc)} for previous four subframes

Definition at line 120 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), and amrnb_decode_init().

previous impulse response filter "impNr": 0 - strong, 1 - medium, 2 - none

Definition at line 129 of file amrnbdec.c.

Referenced by anti_sparseness().

int16_t AMRContext::prev_lsf_r[LP_FILTER_ORDER]

residual LSF vector from previous subframe

Definition at line 103 of file amrnbdec.c.

Referenced by lsf2lsp_3(), lsf2lsp_5(), and lsf2lsp_for_mode12k2().

double AMRContext::prev_lsp_sub4[LP_FILTER_ORDER]

lsp vector for the 4th subframe of the previous frame

Definition at line 105 of file amrnbdec.c.

Referenced by amrnb_decode_init(), lsf2lsp_3(), lsf2lsp_5(), and update_state().

previous fixed gain; used by anti-sparseness processing to determine "onset"

Definition at line 128 of file amrnbdec.c.

Referenced by anti_sparseness().

float AMRContext::samples_in[LP_FILTER_ORDER+AMR_SUBFRAME_SIZE]

floating point samples

Definition at line 137 of file amrnbdec.c.

Referenced by amrnb_decode_frame(), postfilter(), and update_state().

previous input to tilt compensation filter

Definition at line 133 of file amrnbdec.c.

Referenced by postfilter().


The documentation for this struct was generated from the following file:

Generated on Fri Oct 26 02:43:55 2012 for FFmpeg by  doxygen 1.5.8