AAC (Audio Advanced Coding) is an audio data compression format developed by the Fraunhofer Institute in cooperation with AT, Nokia, Sony and Dolby.
As with the MP3 format, it's a form of compression which removes some of the audio data, called "lossy compression". This means eliminating some of the audio data (non-audible frequencies, for example) in order to get the best compression rate possible, while producing an output file which sounds as close as possible to the original data.
The AAC format corresponds to the international standard "ISO/IEC 13818-7", as an extension of MPEG-2, a standard created by the Moving Pictures Expert Group (MPEG).
AAC use variable bite-rare encoding (VBR for short), a method of encoding which modifies the number of bits used per second to encode sound data, depending on how complicated the audio stream is at a given moment. The algorithm used is higher-performing that that of MP3s, which produces better sound quality for small files, while requiring fewer system resources for encoding and decoding.
In contrast to the maximum of two channels (stereo) supported by the MP3 format, AAC enables polyphonic sound with up to 48 channels. The AAC format also offers sampling frequencies ranging from 8 Hz to 96.0 kHz, as opposed to 16 to 48 kHz for mp3s.
The Ogg Vorbis format is not compatible with the MP3 format, which means that a user must use an audio player which supports the format or install a specific codec in order to be able to play Ogg Vorbis files.