Algebraic code-excited linear prediction


Algebraic code-excited linear prediction is a patented speech coding algorithm by VoiceAge Corporation in which a limited set of pulses is distributed as excitation to a linear prediction filter. It is a linear predictive coding algorithm that is based on the code-excited linear prediction method and has an algebraic structure.
The ACELP method is widely employed in current speech coding standards such as AMR, EFR, AMR-WB, VMR-WB, EVRC, EVRC-B, SMV, TETRA, PCS 1900, MPEG-4 CELP and ITU-T G-series standards G.729, G.729.1 and G.723.1. The ACELP algorithm is also used in the proprietary ACELP.net codec.
ACELP is a patented technology and registered trademark of in Canada and/or other countries and was developed in 1989 by the researchers at the Université de Sherbrooke in Canada.

Features

The main advantage of ACELP is that the algebraic codebook it uses can be made very large without running into storage or complexity problems.

Technology

The ACELP algorithm is based on that used in code-excited linear prediction, 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 synthesis filter is given by:
The short-term synthesis filter is given by:
Voiceage have kept very tight control of the product. Audible Inc. use a modified version for their speaking books. It is also licensed conference-calling software, speech compression toys and has become one of
the 3GPP formats. With the patent ending on 9 February 2018, designers of narrow-band speech have the option of ACELP which the customer can optionally pay for now or for standard
usage after the patent expires.