An encoder which encodes a voice in accordance with LD-CELP (low delay code excitation linear prediction encoding) of the ITU-T Recommendation G.728 as a method wherein, when a vibration wave such as a voice signal is encoded by vector quantization, the code is secretly combined with another data. The encoder stores dividing key data kidx by which 128 types of representative vector data (waveform codes) yj; j = 0, 1, ..., 127 are labeled with 0 or 1 in the order from the uppermost bit. If the bit to be combined is "0", the vectors are quantized by using only the waveform codes yj corresponding to the bit "0" of the dividing key data kidx as the selection objects. If the bit to be combined is "1", the vectors are quantized by using only the waveform codes yj corresponding to the bit "1" of the dividing key data kidx as the selection objects. Thus, the outputted voice code is combined with another datum bit.