4 C code structure

26.2733GPPANSI-C code for the fixed-point Extended Adaptive Multi-Rate - Wideband (AMR-WB+) speech codecRelease 17TS

This clause gives an overview of the structure of the bit‑exact C code and provides an overview of the contents and organization of the C code attached to the present document.

The C code has been verified on the following systems:

– IBM PC/AT compatible computers with Windows 2000 SP4 and Microsoft Visual C++ v.6.0 compiler.

ANSI‑C was selected as the programming language because portability was desirable.

4.1 Contents of the C source code

The C code distribution has the files divided in six different directories, all present in the directory c-code. The directories are: common_fx, decoder_fx, encoder_fx, lib_amr_fx include_fx and basic_op. The distributed files with suffix "c" contain the source code and the files with suffix "h" are the header files.

Project and workspace files are provided in the directory MSVC.

4.2 Program execution

The Extended Adaptive Multi-Rate Wideband codec is implemented in two programs:

– (encoder) audio encoder;

– (decoder) audio decoder.

The programs should be called like:

– encoder [encoder options] -if <audio input file> -of <parameter file>;

– decoder [decoder options] -if <parameter file> -of <audio output file>.

The input files contain one or two channels of 16-bit linear encoded PCM audio samples stored in the wav file format and the parameter files contain encoded audio data and some additional flags.

The encoder and decoder options will be explained by running the applications without input arguments. Detailed information explaining the use of the AMR-WB+ encoder and decoder programs is provided in the user’s guide in Annex A of [1]. See also the file readme.txt for condensed information on how to run the encoder and decoder programs.

4.3 Code hierarchy

Tables 1 and 2 are call graphs that show the functions used in the audio codec.

Each column represents a call level and each cell a function. The functions contain calls to the functions in rightwards neighbouring cells. The time order in the call graphs is from the top downwards as the processing of a frame advances. All standard C functions: printf(), fwrite(), etc. have been omitted. Also, no basic operations (add(), L_add(), mac(), etc.) or double precision extended operations (e.g. L_Extract(), Copy() or Set_zero()) appear in the graphs. The initialization of the static RAM (i.e. calling the _init functions) is also omitted.

The basic operations are not counted as extending the depth, therefore the deepest level in this software is level 6.

Table 1: Encoder call structure

coder_amrwb_plus_stereo

Scale_speech_st

Rescale_enc_mem_st

Scale_sig

Decim_12k8_p

Interpol

HP50_12k8_p

Mix_ch

Preemph_scaled

Rescale_enc_mem_st

Coder_lf

AutocorrPlus

Lag_window_p

Levinson

Az_isp

Chebps2

Int_lpc

Isp_Az

Get_isp_pol

Find_wsp

Weight_a

Residu2

Deemph

LP_Decim2

Scale_sig

scale_mem_Hp_wsp

Pitch_med_ol

Hp_wsp

Isp_isf

Qpisf_2s_46b

VQ_stage1

Sub_VQ

Dpisf_2s_46b

Reorder_isf

coder_acelp

Scale_sig

Residu

Weight_a

Syn_filt

Deemph2

Preemph

Syn_filt_s

Pitch_fr4

Norm_Corr

Convolve

Interpol_4

Pred_lt4

Convolve

G_pitch

Updt_tar

Pit_shrp

cor_h_x

ACELP_4t64_fx

cor_h_vec

search_ixiy

Quant_2p_2N1

Quant_1p_N1

Quant_3p_3N1

Quant_2p_2N1

Quant_4p_4N

Quant_4p_4N1

Quant_2p_2N1

Quant_5p_5N

Quant_3p_3N1

Quant_2p_2N1

Quant_6p_6N_2

Quant_5p_5N

Quant_3p_3N1

Quant_2p_2N1

Quant_4p_4N

Quant_4p_4N1

Quant_2p_2N1

Quant_3p_3N1

Quant_2p_2N1

Quant_2p_2N1

Quant_1p_N1

Q_gain2_plus

Segsnr

Coder_tcx

Cos_window

Weight_a

Residu2

Deemph

fft9_fx

init_r_fft_fx

r_fft_fx

FFT_reorder

Adap_low_freq_emph

SpPeak1k6

AVQ_Cod

RE8_PPV_FX

Nearest_neighbor_2D8

Scale_tcx_ifft

Scale_sig

Adap_low_freq_deemph

SpPeak1k6

Deemph1k6

Ifft_reorder

ifft9_fx

r_fft_fx

Get_gain

Q_gain_tcx

Preemph

Scale_sig

Syn_filt_s

Scale_speech

Rescale_enc_mem_hf

coder_hf

AutocorrPlus

Lag_window_p

Levinson_p

Az_isp_p

Chebps2

Int_lpc

Isp_Az

Get_isp_pol

Isp_isf

Q_isf_hf

Sub_VQ

Reorder_isf

Isf_isp

Match_gain_6k4

Residu

Syn_filt

Scale_sig

Int_gain

Residu2

Syn_filt_s

E_LPC_a_weight

Weight_a

Q_gain_hf

Band_split_taligned_2k

Decim_2k

Interpol_st

Oversamp_2k

Interpol_st

Coder_stereo_x

Cod_hi_stereo

Residu2

Compute_exc_side

Compute_cross_corr_vector

Cholsolc

Cholsolc_repair

Smooth_ener_filter

quant_filt

Pmsvq2

Msvq2

M_cbcod

Fir_filt

Compute_gain_match

Quant_gain

Cod_tcx_stereo

Ctcx_stereo

Cos_window

Windowing

Comp_gain_shap_cod

Apply_gain_shap

Get_gain

Q_gain_pan

Compute_xn_target

Fft3

r_fft_fx

FFT_reorder

Adap_low_freq_emph

SpPeak1k6

AVQ_Cod

RE8_PPV_FX

Nearest_neighbor_2D8

Scale_tcx_ifft

Scale_sig

Adap_low_freq_deemph

SpPeak1k6

Deemph1k6

Ifft_reorder

Ifft3

r_fft_fx

Q_gain_tcx

Apply_tcx_overlap

Segsnr

Int2bin

Enc_prm

Int2bin

AVQ_encmux

split_idx_noovf

sort(avq_cod.c)

RE8_cod

RE8_vor

Re8_identify_absolute

Leader

Re8_coord

Re8_k2y

RE8_PPV_FX

Nearest_neighbor_2D8

Re8_compute_base_index

Re8_compute_rank_of_permutation_and_s

calc_bits

writ_all_nq

calc_bits

writ_all_i

init_pos_i_ovf

chk_ovf

chk_ovf

writ_I

writ_k

writ_ovf

Unpack4bits

Int2bin

Enc_prm_stereo_x

Iint2bin

AVQ_encmux

split_idx_noovf

sort(avq_cod.c)

RE8_cod

RE8_vor

re8_identify_absolute

Leader

re8_coord

re8_k2y

RE8_PPV_FX

nearest_neighbor_2D8

re8_compute_base_index

re8_compute_rank_of_permutation_and_s

calc_bits

writ_all_nq

calc_bits

writ_all_i

init_pos_i_ovf

chk_ovf

chk_ovf

writ_I

writ_ovf

writ_k

unpack4bits_d

int2bin

Enc_prm_hf

int2bin

Table 2: Decoder call structure

Decoder_amrwb_plus

Bin2int

Dec_prm

Bin2int

Pack4bits

Bin2int

Dec_prm_stereo_x

Bin2int

Pack4bits_d

Bin2int

AVQ_Demuxdec

Read_all_nq

Read_nq

Read_all_i

Init_pos_i_ovf

Chk_ovf

Split_n

Chk_ovf

Read_I

Read_ovf

Read_k

RE8_Dec

Re8_decode_base_index

Re8_decode_rank_of_permutation

Re8_k2y

RE8_PPV_FX

Nearest_neighbor_2D8

Dec_prm_hf

Bin2int

Decoder_amrwb_plus_1

Decoder_lf

Dpisf_2s_46b_2

Reorder_isf

Isf_isp

Int_lpc

Isp_Az

Get_isp_pol

decoder_tcx

AVQ_Demuxdec

Read_all_nq

Read_nq

Read_all_i

Init_pos_i_ovf

Chk_ovf

Split_n

Chk_ovf

Read_I

Read_k

Read_ovf

RE8_dec

Re8_decode_base_index

Re8_decode_rank_of_permutation

Re8_k2y

RE8_PPV_FX

Nearest_neighbor_2D8

Scale_tcx_ifft

Scale_sig

NoiseFill

Rnd_ph16

Random

Adap_low_freq_deemph

SpPeak1k6

Deemph1k6

Find_mpitch

Ifft_reorder

ifft9_fx

r_fft_fx

Cos_window

D_gain_tcx

Scale_mem_tcx

Scale_sig

Preemph

Weight_a

Syn_filt_s

Residu2

Decoder_acelp

Pred_lt4

DEC_ACELP_4t64_fx

Dec_1p_N1

Dec_2p_2N1

Dec_3p_3N1

Dec_2p_2N1

Dec_1p_N1

Dec_4p_4N

Dec_4p_4N1

Dec_1p_N1

Dec_1p_N1

Dec_2p_2N1

Dec_3p_3N1

Dec_2p_2N1

Dec_1p_N1

Dec_5p_5N

Dec_2p_2N1

Dec_3p_3N1

Dec_2p_2N1

Dec_1p_N1

Dec_6p_6N_2

Dec_1p_N1

Dec_2p_2N1

Dec_3p_3N1

Dec_2p_2N1

Dec_1p_N1

Dec_4p_4N

Dec_4p_4N1

Dec_1p_N1

Dec_5p_5N

Dec_2p_2N1

Dec_3p_3N1

Dec_2p_2N1

Dec_1p_N1

add_pulses

Preemph

Pit_shrp

D_gain2_plus

Scale_exc

Scale_sig

Scale_sig

voice_factor

Rescale_mem

Scale_sig

Syn_filt_s

Weight_a

Residu

Deemph2

Updt_mem_q

Scale_mem2

Scale_sig

Deemph

Bass_postfilter

Scale_sig

Short_pitch_tracker

Decoder_hf

D_isf_hf

Reorder_isf

Isf_isp

Int_lpc

Isp_Az

Get_isp_pol

Match_gain_6k4

Residu

Syn_filt

Scale_sig

Int_gain

D_gain_chan

D_gain_hf

Soft_exc_hf

Syn_filt_s

Smooth_ener_hf

Delay

Oversamp_12k8

Interpol

Decoder_stereo_x

Band_split_taligned_2k

Decim_2k

Interpol_st

Oversamp_2k

Interpol_st

Dec_tcx_stereo

Dtcx_stereo

Cos_window

Scale_tcx_ifft

Scale_sig

Windowing

Comp_gain_shap

adap_lo_freq_deemph

SpPeak1k6

Deemph1k6

Ifft_reorder

Ifft3

ifft_rel

Balance

D_Balance

D_gain_tcx

Apply_xnq_gain2

Apply_wien_filt

Crosscorr_2

Glev_s

Apply_tcx_overlap

Get_alpha

Ch_sep

Dec_hi_stereo

Dec_filt

Dec_gain

Residu

Fir_filt

Get_exc_win

Get_exc

Syn_filt_s

Delay

Band_join_2k

Interpol_st

HP50_12k8

Oversamp_12k8

Interpol

4.5 Variables, constants and tables

The data types of variables and tables used in the fixed point implementation are signed integers in 2’s complement representation, defined by:

– Word16 16 bit variable;

– Word32 32 bit variable.

4.5.1 Description of fixed tables used in the C-code

This clause contains a listing of all fixed tables sorted by source file name and table name. All table data is declared as Word16.

Table 3: Encoder fixed tables

Format

Table name

Size

Description

Format

Word16

MonoRate

54

Predefined mono rate

Q6/Q0/Q0

Word16

StereoRate

81

Predefined stereo rate

Q6/Q0/Q0

Word16

NBITS_CORE_AMR_WB_FX

9 miMode_fx

AMR WB Core bits

Q0

Word16

miMode_fx

48

Stereo Mode Index

Q0

Word16

isfIndex_fx

14

Internal sampling Frequency

Q0

Word16

NBITS_CORE_FX

8

Core bit-rates

Q0

Word16

t_sinFxS4

1440

FFT Sine&Cos table

Q13

Word16

Filter_32k

61

FIR table for decimation/oversampling

Q13

Word16

Filter_32k_hf

61

FIR table for decimation/oversampling

Q13

Word16

Filter_32k_7k

61

FIR table for decimation/oversampling

Q13

Word16

Filter_48k

185

FIR table for decimation/oversampling

Q13

Word16

Filter_48k_hf

185

FIR table for decimation/oversampling

Q13

Word16

Filter_8k

61

FIR table for decimation/oversampling

Q13

Word16

Isf_init

16

Initial ISF memory

Word16

Isp_init

16

Initial ISP memory

Q15

Word16

Isp_init_HF

8

HF Initial ISP memory

Q15

Word16

Mean_isf

16

Means of ISFs

Word16

Dico1_isf

2304

1st stage codebook, isf0 to isf8

Word16

Dico2_isf

1792

1st stage codebook, isf9 to isf15

Word16

Dico21_isf

192

2nd stage codebook, isf2_0 to isf 2_2

Word16

Dico22_isf

384

2nd stage codebook, isf2_3 to isf 2_5

Word16

Dico23_isf

384

2nd stage codebook, isf2_6 to isf 2_8

Word16

Dico24_isf

96

2nd stage codebook, isf2_9 to isf 2_11

Word16

Dico25_isf

128

2nd stage codebook, isf2_12 to isf 2_15

Word16

Dico21_isf_36b

640

1st stage codebook, (36b) split 1

Word16

Dico22_isf_36b

512

1st stage codebook, (36b) split 2

Word16

Dico23_isf_36b

448

1st stage codebook, (36b) split 3

Word16

Dico_gain_hf

512

Quantization table for one-stage HF gain

Q8

Word16

Mean_isf_hf_12k8

8

Means of ISFs (full band)

Word16

Dico1_isf_hf_12k8

32

1st stage isf codebook (full band)

Word16

Mean_isf_hf_low_rate

8

Means of isfs

Word16

Dico1_isf_hf_low_rate

32

1st stage isf codebook

Word16

Dico2_isf_hf

1024

2nd stage isf codebook

Word16

Filt_lp

13

Low-pass fir filter for bass post filter

Q15

Word16

Sin20

20

Random phase

Q15

Word16

Inter4_2

128

1/4 resolution interpolation filter

Q14

Word16

VadFiltBandFreqs

12

Open-loop classifier

Q0

Word16

Bw_inv

12

Open-loop classifier

Q22

Word16

Lwg

8

Open-loop claissifier

Q15

Word16

Gain_hf_ramp

64

HF gain ramp for wb->wb+ switiching

Q15

Word16

Inter2_coef

12

Filter coefficients for band join/split

Q13

Word16

Filter_LP180

2341

Filter for 48 kHz interpolation

Q14

Word16

StereoNbits_FX

18

Stereo bit-rates

Q0

Word16

Filter_2k_fxQ14_32

321

2k decimation filter

Q14

Word16

Filter_2k_fxQ14_5

321

2k decimation filter

Q14

Word16

cb_filt_hi_mean_fx

9

Average filter

Q14

Word16

Filt_hi_mscb_4a_fx

16*9

Q14

Word16

Filt_hi_mscb_7a_fx

16*9

Q14

Word16

Filt_hi_mscb_7b_fx

8*9

Q14

Word16

Cb_gain_hi_mean_fx

2

Average gain vector

Q10

Word16

Gain_hi_mscb_2a_fx

4*2

Q10

Word16

Gain_hi_mscb_5a_fx

32*2

Q10

Word16

Dico1_isf_hf_high_rate

32

1st stage isf codebook

Word16

Mean_isf_hf_high_rate

8

Means of isfs

Word16

Filter_LP45

586

Filter for 48 kHz interpolation

Q14

Word16

t_qua_gain6b

128

Gain pitch and gain code

Q14/Q11

Word16 t_qua_gain7b

t_qua_gain7b

256

Gain pitch and gain code

Q14/Q11

Word16

Overlap_wind

63

Overlap window

Q15

Word16

Cos_wind

128

Cos window

Q15

Word16

Cos_wind_LR

224

Cos _window (Low rate)

Q15

Word16

TXV

31

Arctan piece table

Q15

Word16

Len_tbl

6

Inverse length

Q15

Word16

interpol_frac4

4

Interpolation Window 4 sub-frame

Q15

Word16

interpol_frac8

8

Interpolation Window 8 sub-frame

Q15

Word16

interpol_frac16

16

Interpolation Window 16 sub-frame

Q15

Word16

size_filt_hi_msvq_4_fx

16

Stereo param

Word16

*cbs_filt_hi_msvq_4_fx

16

Stereo param

PMSQ_fx

filt_hi_pmsvq4_fx

Word16

size_filt_hi_msvq_7_fx

16

Stereo param

Word16

*cbs_filt_hi_msvq_7_fx

16

Stereo param

PMSQ_fx

filt_hi_pmsvq7_fx

Word16

Size_gain_hi_msvq_2_fx

16

Stereo param

Word16

*cbs_gain_hi_msvq_2_fx

16

Stereo param

PMSQ_fx

gain_hi_pmsvq2_fx

Word16

size_gain_hi_msvq_5_fx

16

Stereo param

Word16

*cbs_gain_hi_msvq_5_fx

16

Stereo param

PMSQ_fx

gain_hi_pmsvq5_fx

Table 4: Decoder fixed tables

Format

Table name

Size

Description

Same as encoder

Same as encoder

4.5.2 Static variables used in the C-code

In this clause two tables that specify the static variables for the encoder and decoder respectively are shown. All static variables are declared within a C struct.

Table 6: Encoder static variables

Struct name

Type

Variable

Type[Length]

Description

Coder_StState_fx

Word16

mem_decim

1608

speech decimated filter memory

Word16

decim_frac

1

Fractional decimation factor

Word16

mem_sig_in

6

hp filter memory

Word16

mem_preemph

1

speech preemphasis filter mem

Word16

mem_decim_hf

46

HF filter memory

Word16

old_speech_hf

528

HF old speech vector

Word16

past_q_isf_hf

8

HF past quantized isf

Word16

ispold_hf

8

HF old isp

Word16

ispold_q_hf

8

HF quantized old isp

Word16

old_gain;

1

HF old gain match

Word16

mem_hf1

8

HF memory for gain 1

Word16

mem_hf2

8

HF memory for gain 2

Word16

mem_hf3

8

HF memory for gain 3

Word16

old_exc

375

old excitation

Word16

Q_sp_hf

1

Scaling hf speech

Word16

OldQ_sp_hf

2

old scaling hf speech

Word16*

mean_isf_hf

1

isf codebook mean

Word16*

dico1_isf_hf

1

isf codebook first stage

Coder_State_Plus_fx

Coder_StState_fx

Left

2617

state for left channel

Coder_StState_fx

Right

2617

state for right channel

Word16

old_chan

528

old left signal

Word16

old_chan_2k

140

old left signal 2 kHz sampl. rate

Word16

old_chan_hi

448

old left signal HB

Word16

old_speech_2k

140

old mono signal 2 kHz sampl. rate

Word16

old_speech_hi

448

old mono signal HB

Word16

old_speech_pe

528

past pre-emphasised mono

Word16

old_wh

9

past weighted filter

Word16

old_wh_q

9

past quantized weighted filter

Word16

old_gm_gain

2

past gain matching

Word16

old_exc_mono

9

past mono excitation

Word16

filt_energy_threshold

1

filter energy thershold

Word16

w_window

64

weighting window

PMSVQ_fx*

*filt_hi_pmsvq

1

MSVQ quantizer

PMSVQ_fx*

*gain_hi_pmsvq

1

MSVQ quantizer

Word16

mem_stereo_ovlp_size

1

past stereo overlap size

Word16

mem_stereo_ovlp

32

past stereo overlap

NCLASSDATA

*stClass

1

use case B classifier

VadVars

*vadSt

1

VAD state

Word16

vad_hist

1

VAD history

Word16

old_speech

528

old speech

Word16

old_synth

16

synthesis memory

Word16

past_isfq

16

past isf quantizer

Word16

old_wovlp

128

last tcx overlap

Word16

old_d_wsp

187

Weighted speech vector

Word16

old_exc

392

old excitation vector

Word16

old_mem_wsyn

1

weighted synthesis memory

Word16

old_mem_w0

1

weighted speech memory

Word16

old_mem_xnq

1

quantized target memory

Word16

old_ovlp_size

1

last tcx overlap size

Word16

Isfold

16

old isf frequency domain

Word16

Ispold

16

old isp

Word16

ispold_q

16

quantized old isp

Word16

mem_wsp

1

wsp vector mem

Word16

mem_lp_decim2

3

wsp decimator filter mem

Word16

ada_w

1

open loop LTP

Word16

ol_gain

1

open loop LTP

Word16

ol_wght_flg

1

open loop LTP

Word16

old_ol_lag

5

past openloop lag

Word16

old_T0_med

1

past pitch

Word16

hp_old_wsp

699

past HP weighted speech

Word16

hp_ol_ltp_mem

9

past HP openloop long term prediction

Word16

window

512

LP analysis window

Word16

SwitchFlagPlusToWB

1

flag for switching to AMR-WB

Word16

mem_gain_code

4

past code gain

Word16

prev_mod

1

past frame type

Word16

Q_sp

1

Scaling of speech

Word16

OldQ_sp

1

Old scaling of speech

Word16

i_offset

1

Word16

pit_max

1

Mem of pit_max

Word16

lev_mem

18

Levinson durbin memory

Word16

old_wsp_max

4

Weight speech scaling memory

Word16

old_wsp_shift

1

limit dynamic at 12 bits

Word16

scale_fac

1

scaling factor (preemph speech)

Word16

Q_new

1

scaling factor of speech

Word16

Q_max

2

Q_new limitation

Word16

OldQ_sp_deci

2

Q_new memory

Word16

Q_exc

1

excitation scaling

Word16

Old_Qexc

1

excitation scaling memory

Word16

LastQMode

1

Last subfr mode (acelp/tcx)

Encoder Config

Word16

mode

1

AMR_WB core mode: 0..8

Word16

extension

1

0=AMRWB, 1=WB+

Word16

st_mode

1

stereo mode

Word16

fscale

1

frequency scaling

Word16

use_case_mode

1

use case (for AMRWB+ only)

Word16

allow_dtx

1

dtx (for AMRWB only)

Word16

FileFormat

1

3gp or raw

Word16

mode_index

1

index of wb+ mode used

Word16

fscale_index

1

index of internal frequency sampling

Word16

bc

1

Backward compatible file format

Table 7: Decoder static variables

Struct name

Type

Variable

Type[Length]

Description

Decoder_StState

Word16

wmem_oversamp

72

Memory oversampling

Word16

wover_frac

1

Fractional overcloking factor

Word16

wmem_oversamp_hf

24

memory

Word16

wpast_q_isf_hf

8

HF past quantized isf

Word16

wpast_q_isf_hf_other

8

HF past quantized isf for the other

channel when mono decoding stereo

Word16

wpast_q_gain_hf

1

HF past quantized gain

Word16

wpast_q_gain_hf_other

1

HF past quantized gain for the other

channel when mono decoding stereo

Word16

wold_gain

1

HF old gain match

Word16

wispold_hf

8

HF old isp

Word32

Lthreshold;

1

HF memory for smooth ener

Word16

wmem_syn_hf

8

HF synthesis memory

Word16

mem_d_tcx_fx

96

delay compensation memory

Word16

wmem_d_nonc

64

Non causality delay

Word16

mem_synth_hi

16

High band sunthesis memory

Word16

wmem_sig_out

6

hp filter memory

Word16

wold_synth_hf

512

synch delay memory

Word32

Lp_amp

1

memory for soft exc

Word16*

mean_isf_hf

1

isf codebook mean

Word16*

dico1_isf_hf

1

isf codebook first stage

Word16

Q_synHF

1

scaling of hf synth

Decoder_State_Plus

Decoder_StState

left

833

State for left channel

Decoder_StState

right

833

State for right channel

Word16

mem_left_2k_fx

20

2kHz memory on left chan

Word16

mem_right_2k_fx

20

2kHz memory on right chan

Word16

mem_left_hi_fx

64

HB memory left channel

Word16

mem_right_hi_fx

64

HB memory right channel

Word16

my_old_synth_2k_fx

35

old 2kHz synthesis

Word16

my_old_synth_hi_fx

128

old HB synthesis

Word16

my_old_synth_fx

148

old stereo synth

Word16

old_AqLF_fx

85

old quantized LPC

Word16

old_wh_fx

9

old decoded filter

Word16

old_wh2_fx

9

old decoded filter 2

Word16

old_exc_mono_fx

9

old mono excitation

Word16

old_gain_left_fx

4

old gain on left chan

Word16

old_gain_right_fx

4

old gain on right chan

Word16

old_wh_q_fx

9

past quantized filter

Word16

old_gm_gain_fx

2

past gain matching

Word16

W_window

64

weighted synthesis window

PMSVQ

*filt_hi_pmsvq_fx

1

past MSVQ filter

PMSVQ

*gain_hi_pmsvq_fx

1

past MSVQ gain

Word16

mem_stereo_ovlp_size_fx

1

past stereo overlap size

Word16

mem_stereo_ovlp_fx

32

past stereo overlap

Word16

last_stereo_mode

1

past stereo mode

Word16

side_rms_fx

1

side signal RMS

Word16

H_fx

9

current filter

Word16

wold_xri

1 148

old spectral coefficeints

Word16

last_mode

1

last mode in previous 80ms frame

Word16

wmem_sig_out

6

hp50 filter memory for synthesis

Word16

wmem_deemph

1

speech deemph filter memory

Word16

prev_lpc_lost

1

previous lpc is lost when = 1

Word16

wold_synth

16

synthesis memory

Word16

wold_exc

392

old excitation vector

Word16

wisfold

16

old isf (frequency domain)

Word16

wispold

16

old isp (immittance spectral pairs)

Word16

wpast_isfq

16

past isf quantizer

Word16

wovlp

128

last weighted synthesis for overlap

Word16

ovlp_size

1

overlap size

Word16

wisf_buf

51

old isf (for frame recovery)

Word16

wold_T0

1

old pitch value (for frame recovery)

Word16

wold_T0_frac

1

old pitch value (for frame recovery)

Word16

seed_ace

1

seed memory (for random function)

Word16

wmem_wsyn

1

TCX synthesis memory

Word16

seed_tcx

1

seed memory (for random function)

Word16

wwsyn_rms

1

rms value of weighted synthesis

Word16

wpast_gpit

1

past gain of pitch (for frame recovery)

Word32

Lpast_gcode

1

past gain of code (for frame recovery)

Word16

pitch_tcx

1

for bfi

Word32

L_gc_threshold

1

GC threshold

Word16

wold_synth_pf

503

Bass post-filter: old synthesis

Word16

wold_noise_pf

24

bass post-filter: noise memory

Word16

wold_T_pf

2

bass post-filter: old pitch

Word16

wold_gain_pf

2

Bass post-filter: old pitch gain

Word16

*mean_isf_hf

1

HF isf codebook in-use

Word16

*dico1_isf_hf

1

HF isf codebook in-use

Word16

wmem_gain_code

4

past code gain

Word16

wmem_lpc_hf

9

past HF lpc filter

Word16

wmem_gain_hf

1

past HF gain

Word16

wramp_state

1

ramp state

Word16

cp_old_synth

16

old synthesis switching memory

Word16

Q_old

1

Old scaling

Word16

Q_exc

1

excitation scaling

Word16

Q_syn

1

synthesis scaling

Word16

Old_Q_syn

1

Old synthesis scaling

Word16

Old_Q_exc

1

Old excitation scaling

Word16

prev_Q_syn

1

Limitation on synthesis scaling

Word16

mem_syn2

16

Switching synthesis memory

Word16

Old_Qxnq

1

Old xnq scaling

Word16

Old_QxnqMax

1

Old maximum xnq scaling

Word16

Old_Qxri

1

Old xri scaling

Word16

Old_bpf_scale

1

noise buf scaling

Word16

mem_subfr_q

7

subfr maximum excitation scaling

Word16

old_subfr_q

16

subfr true excitation scaling

Word16

old_syn_q

16

subfr true synthesis scaling

Word16

i_offset

1

offset memory

DecoderConfig

Word16

mode

1

AMR_WB core modes

Word16

extension

1

0 = AMR_WB 1 = WB+

Word16

st_mode

1

Stereo modes

Word16

fscale

1

Internal Frequency scaling

Word32

fs

1

Sampling rate

Word32

mono_dec_stereo

1

decode mono a stereo bitstream

Word32

limiter_on

1

Limite clipping

Word16

Fileformat

1

File format used (raw/3gp)

Word16

fer_sim

1

Frame erasure simulation