DEVICE AND METHOD FOR CELP TYPE VOICE DECODING

摘要

PROBLEM TO BE SOLVED: To obtain a synthesized voice of higher quality by providing a diffusion pattern storage and selection part, a pulse diffusion part which convolutes a diffusion pattern selected by the selection part and a pulse vector, etc., and supplying a pulse vector by an algebraic code book. SOLUTION: The pulse vector diffusion part 313 generates a diffusion vector from a diffusion pattern received from a diffusion pattern storage and selection part 311 and a pulse vector received from a pulse vector generation part 312 through specific superposing operation and outputs it to a diffusion vector addition part 314. The diffusion vector addition part 314 adds diffusion vectors of respective channels generated by the pulse vector diffusion part 313 to generate a noise code vector. Then an adaptive code gain and a noise code gain corresponding to a weight code number are read out of a weight code book 305 and a pulse vector is supplied by an adaptive code book 303 being an algebraic code book to an adaptive code vector weighting part 306.