Gain-shape vector quantization apparatus

A gain-shape vector quantization apparatus for compressing the data of voice signals, comprising a code book portion (1) which is constituted by a plurality of shape vectors and produces a plurality of selected shape vectors, a plurality of variable gain circuits (2₁ - 2n) that impart gains to the shape vectors produced from said code book (1), a plurality of synthesis filters (3₁ - 3n) that regenerate signals from the outputs of said variable gain circuits (2₁ - 2n), an adder for adding the signals regenerated by said synthesis filters (3₁ - 3n), and an evaluation unit (5) which produces an index to select a plurality of shape vectors in the code book portion in order to minimize the error between the output of said adder and an input speech and further produces gain adjusting signals for the variable gain circuits.