Editing Algebraic code-excited linear prediction

{{Short description|Speech coding standard}}
{{Use dmy dates|date=February 2019}}

'''Algebraic code-excited linear prediction''' ('''ACELP''') is a [[speech coding]] algorithm in which a limited set of pulses is distributed as excitation to a [[Linear predictive coding|linear prediction]] filter. It is a [[linear predictive coding]] (LPC) algorithm that is based on the [[code-excited linear prediction]] (CELP) method and has an [[algebra]]ic structure. ACELP was developed in 1989 by the researchers at the [[Université de Sherbrooke]] in [[Canada]].<ref>{{Cite web|url=https://www.usherbrooke.ca/recherche/fr/partenariats/pour-les-entreprises/transfert-de-technologie/exemples-de-transferts/technologie-de-compression-de-la-parole-acelpr/|title=Transfer of technology}}</ref>

The ACELP method is widely employed in current speech coding standards such as [[Adaptive Multi-Rate|AMR]], [[Enhanced Full Rate|EFR]], [[AMR-WB]] (G.722.2), [[VMR-WB]], [[EVRC]], [[EVRC-B]], [[Selectable Mode Vocoder|SMV]], [[TETRA]], PCS 1900, [[MPEG-4]] CELP and [[ITU-T]] G-series standards [[G.729]], [[G.729.1]] (first coding stage) and [[G.723.1]].<ref>[https://web.archive.org/web/20071014224013/http://www.voiceage.com/image/ACELPMAP.jpg ACELP map], VoiceAge Corporation, Archive.org</ref><ref>{{Cite web|url=http://www.voiceage.com/relatedstandards.php|title=Related Standards Specifications|date=14 October 2007|archive-url=https://web.archive.org/web/20071014095716/http://www.voiceage.com/relatedstandards.php |archive-date=14 October 2007 }}</ref><ref name="archive-voiceage-technologies">{{cite web |author=VoiceAge Corporation |url=http://www.voiceage.com/technologies.php |title=Codec Technologies |date=2007-10-13 |access-date=2009-09-20 |archive-url=https://web.archive.org/web/20071013162836/http://www.voiceage.com/technologies.php |archive-date=2007-10-13}}</ref><ref name="voiceage-technologies">{{cite web|author=VoiceAge Corporation |publisher=VoiceAge Corporation |url=http://www.voiceage.com/technologies.php |title=Codec Technologies |access-date=2009-09-20 |archive-url=https://web.archive.org/web/20091018020920/http://www.voiceage.com/technologies.php |archive-date=18 October 2009 |url-status=dead }}</ref> The ACELP algorithm is also used in the proprietary ACELP.net codec.<ref>{{cite web |url=http://www.voiceage.com/acelpnet.php |title=ACELP.net — Beyond the Standards |author=VoiceAge Corporation |access-date=2010-01-03 |archive-url=https://web.archive.org/web/20071014035312/http://www.voiceage.com/acelpnet.php |archive-date=2007-10-14}}</ref>  [[Audible Inc.]] use a modified version for their speaking books. It is also used in conference-calling software, speech compression tools and has become one of the [[3GPP]] formats.

The ACELP patent expired in 2018 and is now royalty-free.<ref>{{US patent reference | number=5717825 | issue-date=10 February 1998 | inventor= | title=Algebraic code-excited linear prediction speech coding method}}</ref>

==Features==
The main advantage of ACELP is that the [[algebra]]ic codebook it uses can be made very large (> 50 bits) without running into storage ([[random-access memory|RAM]]/[[Read-only memory|ROM]]) or complexity ([[CPU]] time) problems.

==Technology==
The ACELP algorithm is based on that used in [[code-excited linear prediction]] (CELP), but ACELP codebooks have a specific algebraic structure imposed upon them.

A 16-bit algebraic codebook shall be used in the innovative codebook search, the aim of which is to find the best innovation and gain parameters. The innovation vector contains, at most, four non-zero pulses.

In ACELP, a block of ''N'' speech samples is synthesized by filtering an appropriate innovation sequence from a codebook, scaled by a gain factor ''g'' ''c'', through two time-varying filters.

The long-term (pitch) synthesis filter is given by:

:<math>\frac1{B(z)} = \frac1{ 1 - g_p z^{-T} }</math>

The short-term synthesis filter is given by: 
:<math>\frac1{A(z)} = \frac1{ 1 + \sum_{i=1}^P a_i z^{-i} }</math>

==References==
{{Reflist}}

{{Compression Methods}}

[[Category:Speech codecs]]
[[Category:Data compression]]