AAC technique involves exploiting two primary coding strategies in order to minimize the amount of data required to impart high-quality digital audio. Signal components that are irrelevant are discarded. Redundancies in the coded audio signal are wiped out.

The encoding of digital audio files involves the following steps:

1. Modified discrete cosine transform (MDCT) is used to convert the signal from time-domain to frequency-domain. Filter banks are used to convert an exact number of time samples into frequency samples.
2. The frequency domain is quantized using a psychoacoustic model and is then encoded.
3. Appropriate internal error correction codes are applied.
4. The signal is stacked or transmitted.
5. Luhn mod N algorithm is used for each frame to avoid sample corruption

AAC can sample frequencies ranging from 8Hz to 96kHz and up to 48 channels. It is also able to compress audio that contains streams of complex pulses and square waves better than MP3.

AAC is an international standard used by some major companies like Dolby Laboratories Inc., Sony Corp. and Nokia Corp. AAC is also used as a default audio codec for .m4v format by Apple in iTunes Store video files.