The official title is ITU-T Recommendation G.729: “Coding of speech at 8 kbit/s using Conjugate Structure Algebraic Code-Excited Linear Prediction (CS-ACELP).”

In simple terms, G.729 is a highly detailed set of technical instructions—an algorithm—that defines an extremely efficient method for converting human speech into a compact digital stream of data. Its core specification, and the reason for its widespread use, is its ability to code high-quality speech at a very low data rate: just 8 kilobits per second (kbit/s). This efficiency is critical for saving bandwidth on busy digital networks.

That title is a mouthful, so let’s break down the core technology it describes: CS-ACELP.