When an input speech (5) is separated into a spectrum-envelope information and an excitation signal, and the excitation signal is encoded at each frame based on a plurality of excitation signal positions and a plurality of excitation signal gains, the encoding characteristic is improved according to the present invention. In an excitation signal coding unit (11) for encoding the excitation signal based on the plurality of excitation signal positions and the plurality of excitation signal gains, a temporary gain calculating unit (40) for calculating a temporary gain for each excitation signal position candidate is provided. A pulse position search unit (41) determines the plurality of excitation signal positions by using the temporary gains. A gain coding unit (12) encodes the excitation signal gain based on the determined excitation signal position.
展开▼
