Common code between the AC-3 and E-AC-3 decoders. More...
|Large enough for maximum possible frame size when the specification limit is ignored. More...|
|static int||ff_eac3_parse_header (AC3DecodeContext *s)|
|Parse the E-AC-3 frame header. More...|
|static void||ff_eac3_decode_transform_coeffs_aht_ch (AC3DecodeContext *s, int ch)|
|Decode mantissas in a single channel for the entire frame. More...|
|static void||ff_eac3_apply_spectral_extension (AC3DecodeContext *s)|
|Apply spectral extension to each channel by copying lower frequency coefficients to higher frequency bins and applying side information to approximate the original high frequency signal. More...|
Common code between the AC-3 and E-AC-3 decoders.
Summary of MDCT Coefficient Grouping: The individual MDCT coefficient indices are often referred to in the (E-)AC-3 specification as frequency bins. These bins are grouped together into subbands of 12 coefficients each. The subbands are grouped together into bands as defined in the bitstream by the band structures, which determine the number of bands and the size of each band. The full spectrum of 256 frequency bins is divided into 1 DC bin + 21 subbands = 253 bins. This system of grouping coefficients is used for channel bandwidth, stereo rematrixing, channel coupling, enhanced coupling, and spectral extension.
+-+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+-+ |1| |12| | [12|12|12|12] | | | | | | | | | | | | |3| +-+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+–+-+ ~~~ ~~~~ ~~~~~~~~~~~~~ ~~~ | | | | | | | 3 unused frequency bins–+ | | | | | +–1 band containing 4 subbands | | | +–1 subband of 12 frequency bins | +–DC frequency bin
Definition in file ac3dec.h.
|#define AC3_OUTPUT_LFEON 8|
|#define SPX_MAX_BANDS 17|
|#define AC3_FRAME_BUFFER_SIZE 32768|
Parse the E-AC-3 frame header.
This parses both the bit stream info and audio frame header.
Referenced by parse_frame_header().
Decode mantissas in a single channel for the entire frame.
This is used when AHT mode is enabled.
Referenced by decode_transform_coeffs_ch().
Apply spectral extension to each channel by copying lower frequency coefficients to higher frequency bins and applying side information to approximate the original high frequency signal.
Referenced by decode_audio_block().