Also known as: AAC, Advanced Audio Coding, HE-AAC, AAC-LC
Advanced Audio Coding (AAC) is the MPEG-2/MPEG-4 perceptual audio codec that succeeded MP3, delivering better quality at the same bitrate.1 Unlike a speech vocoder, which models the human voice, AAC codes full-band music and general audio; it works by transforming blocks of PCM samples and discarding detail a listener cannot hear, in the same perceptual-coding family as the Opus codec.
How it works
AAC splits the audio into overlapping blocks and applies a modified discrete cosine transform (MDCT) to move each block from the time domain into a set of frequency coefficients. A psychoacoustic model then estimates, band by band, how much noise the ear will mask behind louder nearby tones, and the encoder quantizes each band only as finely as audibility requires — spending bits where the ear is sensitive and coarsening or dropping detail where it is not. The quantized coefficients are finally packed with entropy (Huffman) coding, so the output bitstream is both perceptually transparent and small. Because it is a lossy transform codec rather than a source model of speech, AAC handles music, applause, and mixed program material that a low-rate voice coder cannot.
Several profiles trade complexity for efficiency. AAC-LC (Low Complexity) is the baseline. HE-AAC — where “HE” is High-Efficiency — adds Spectral Band Replication (SBR), which reconstructs high frequencies from a compact parametric description instead of coding them outright. HE-AAC v2 adds Parametric Stereo (PS), describing the stereo image with a few side parameters, and reaches broadcast-usable quality near 24–48 kbps — a fraction of what MP3 needs for the same result.
In practice
AAC’s low-bitrate strength made it the audio codec of choice for constrained broadcast links. DAB+ carries HE-AAC v2, which let it pack many more stations into a multiplex than the original DAB’s MP2. DRM (Digital Radio Mondiale) also uses AAC-family coding for shortwave and mediumwave digital audio, and digital television (ATSC and DVB) uses AAC for audio. One notable exception: US HD Radio does not use AAC — it carries its own proprietary HDC codec, a perceptually similar but separate design. Choosing AAC over an outright PCM stream is a large gain in spectral efficiency: the same channel carries far more program content.
Relevance to SDR
An SDR listener tuning a digital broadcast multiplex is, after demodulation and error-correction, left with an AAC (or HDC) elementary stream that a software decoder turns back into audio. Knowing which codec a mode uses tells you which decoder library to reach for — welle.io and similar DAB+ receivers embed an HE-AAC decoder, while HD Radio tools need the HDC path. GopherTrunk’s own decode targets are land-mobile trunking voice (P25, DMR, NXDN, TETRA) that use low-rate speech vocoders, not AAC; AAC is documented here as the broadcast-audio counterpart an SDR user meets outside the trunking world.
Sources
-
Advanced Audio Coding — Wikipedia, for AAC’s history as the MP3 successor, the MDCT and psychoacoustic model, and the AAC-LC / HE-AAC / HE-AAC v2 profiles. ↩