WMAVoiceContext Struct Reference

WMA Voice decoding context. More...


Data Fields

GetBitContext gb
 packet bitreader.
int8_t vbm_tree [25]
 converts VLC codes to frame type
int spillover_bitsize
 number of bits used to specify spillover_nbits in the packet header = ceil(log2(ctx->block_align << 3))
int history_nsamples
 number of samples in history for signal prediction (through ACB)
int do_apf
 whether to apply the averaged projection filter (APF)
int denoise_strength
 strength of denoising in Wiener filter [0-11]
int denoise_tilt_corr
 Whether to apply tilt correction to the Wiener filter coefficients (postfilter).
int dc_level
 Predicted amount of DC noise, based on which a DC removal filter is used.
int lsps
 number of LSPs per frame [10 or 16]
int lsp_q_mode
 defines quantizer defaults [0, 1]
int lsp_def_mode
 defines different sets of LSP defaults [0, 1]
int frame_lsp_bitsize
 size (in bits) of LSPs, when encoded per-frame (independent coding)
int sframe_lsp_bitsize
 size (in bits) of LSPs, when encoded per superframe (residual coding)
int min_pitch_val
 base value for pitch parsing code
int max_pitch_val
 max value + 1 for pitch parsing
int pitch_nbits
 number of bits used to specify the pitch value in the frame header
int block_pitch_nbits
 number of bits used to specify the first block's pitch value
int block_pitch_range
 range of the block pitch
int block_delta_pitch_nbits
 number of bits used to specify the delta pitch between this and the last block's pitch value, used in all but first block
int block_delta_pitch_hrange
 1/2 range of the delta (full range is from -this to +this-1)
uint16_t block_conv_table [4]
 boundaries for block pitch unit/scale conversion
int spillover_nbits
 number of bits of the previous packet's last superframe preceeding this packet's first full superframe (useful for re-synchronization also)
int has_residual_lsps
 if set, superframes contain one set of LSPs that cover all frames, encoded as independent and residual LSPs; if not set, each frame contains its own, fully independent, LSPs
int skip_bits_next
 number of bits to skip at the next call to wmavoice_decode_packet() (since they're part of the previous superframe)
uint8_t sframe_cache [SFRAME_CACHE_MAXSIZE+FF_INPUT_BUFFER_PADDING_SIZE]
 cache for superframe data split over multiple packets
int sframe_cache_size
 set to >0 if we have data from an (incomplete) superframe from a previous packet that spilled over in the current packet; specifies the amount of bits in sframe_cache
PutBitContext pb
 bitstream writer for sframe_cache
double prev_lsps [MAX_LSPS]
 LSPs of the last frame of the previous superframe.
int last_pitch_val
 pitch value of the previous frame
int last_acb_type
 frame type [0-2] of the previous frame
int pitch_diff_sh16
 ((cur_pitch_val - last_pitch_val) << 16) / MAX_FRAMESIZE
float silence_gain
 set for use in blocks if ACB_TYPE_NONE
int aw_idx_is_ext
 whether the AW index was encoded in 8 bits (instead of 6)
int aw_pulse_range
 the range over which aw_pulse_set1() can apply the pulse, relative to the value in aw_first_pulse_off.
int aw_n_pulses [2]
 number of AW-pulses in each block; note that this number can be negative (in which case it basically means "zero")
int aw_first_pulse_off [2]
 index of first sample to which to apply AW-pulses, or -0xff if unset
int aw_next_pulse_off_cache
 the position (relative to start of the second block) at which pulses should start to be positioned, serves as a cache for pitch-adaptive window pulses between blocks
int frame_cntr
 current frame index [0 - 0xFFFE]; is only used for comfort noise in pRNG()
float gain_pred_err [6]
 cache for gain prediction
float excitation_history [MAX_SIGNAL_HISTORY]
 cache of the signal of previous superframes, used as a history for signal generation
float synth_history [MAX_LSPS]
 see excitation_history
RDFTContext rdft
RDFTContext irdft
 contexts for FFT-calculation in the postfilter (for denoise filter)
DCTContext dct
DCTContext dst
 contexts for phase shift (in Hilbert transform, part of postfilter)
float sin [511]
float cos [511]
 8-bit cosine/sine windows over [-pi,pi] range
float postfilter_agc
 gain control memory, used in adaptive_gain_control()
float dcf_mem [2]
 DC filter history.
float zero_exc_pf [MAX_SIGNAL_HISTORY+MAX_SFRAMESIZE]
 zero filter output (i.e.
float denoise_filter_cache [MAX_FRAMESIZE]
int denoise_filter_cache_size
 samples in denoise_filter_cache
float tilted_lpcs_pf [0x80]
 aligned buffer for LPC tilting
float denoise_coeffs_pf [0x80]
 aligned buffer for denoise coefficients
float synth_filter_out_buf [0x80+MAX_LSPS_ALIGN16]
 aligned buffer for postfilter speech synthesis


Detailed Description

WMA Voice decoding context.

Definition at line 128 of file wmavoice.c.


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

Generated on Fri Oct 26 02:36:58 2012 for FFmpeg by  doxygen 1.5.8