Editing Musepack (section)

== Technical details ==
Musepack was developed using the [[MP2 (format)|MP2]] codec as a starting point, but many features have since been added, including:
* subband selectable M/S encoding (as in [[Advanced Audio Coding|AAC]])
* [[Huffman coding]] (as in [[MP3]] and AAC, but more efficient). Since SV8 the bitstream is compressed by highly optimized canonical huffman tables that yields 2% smaller files and faster decoding
* noise substitution techniques (as in ATSC A-52 and [[MPEG-4]] AAC V2)
* pure variable bitrate between 0 and 1300&nbsp;kbit/s (when needed)

The [[psychoacoustics|psychoacoustic]] model of MPC is based on MPEG ISO model 2, but is extended by CVD (clear voice detection). The quantization algorithm of the MPC encoder performs spectral shaping of the noise, called ''[[adaptive noise shaping]]'' (ANS), in order to overcome the low frequency resolution of the [[polyphase quadrature filter]] bands.

MPC uses the [[APEv2 tag]] [[metadata]] container.<ref>{{cite web|url=http://trac.musepack.net/musepack/wiki/SV8Specification |title=SV8 specification |publisher=Musepack.net |date=2009-02-24 |access-date=2020-02-14}}</ref>

Musepack is mainly optimized for transparent encoding at the "--standard" preset (175-185&nbsp;kbit/s). Very few optimisations have been made at lower bitrates (like 128&nbsp;kbit/s). Nevertheless, various [[codec listening test|listening tests]] have been conducted in which Musepack has performed well at both lower and higher bitrates.<ref>{{cite web|url=http://soundexpert.org/encoders-128-kbps |title=Audio quality of encoders at 128 kbit/s |publisher=Soundexpert.org |access-date=2011-04-25}}</ref><ref>{{cite web |url=http://audiotests.free.fr/tests/2004.07/hq1 |title=MPC vs OGG VORBIS vs MP3 at 175 kbps, listening test on non-killer samples |publisher=Audiotests.free.fr |access-date=2011-04-25 |archive-url=https://web.archive.org/web/20110719184141/http://audiotests.free.fr/tests/2004.07/hq1/ |archive-date=2011-07-19 |url-status=dead }}</ref>